My dad got me into the sport, initially taking me and my sister on easy jogs around the block, gradually building up to more challenging routes and distances.

For a long time, my relationship with running had nothing to do with competition.

I ran because I enjoyed it, because it helped clear my mind, and because it was something I did with my family.

To the extent I was interested in getting faster, the metrics I used to track my fitness were rudimentary. I had a regular, uphill route I ran at least once a week, and a basic Casio wristwatch to measure my time.

I had no concept of the distance I was running, nor of my split pace, nor of the total elevation. All I had was a general sense that if I could complete the route in under 30 minutes, I was in pretty good shape.

Over the last few years, I’ve noticed my relationship with running change: a change that started when I downloaded Strava.

With Strava, I can now see detailed metrics for every run I complete, including distance, heart-rate zones, and pace per kilometre. Through the app, others can see these stats too.

On the one hand, seeing my average pace per kilometre has arguably made me a better runner, motivating me to build fitness and bring that number down. At a subconscious level, the public nature of this stat likely pushes me to try a little harder than I otherwise would.

But what happens when 00:00/KM stops being just a metric and starts to become a target?

If maximizing my pace per kilometre becomes my primary goal, this incentivizes me to avoid more challenging uphill routes. It also disincentivizes me to complete longer runs at an easy pace: the kind of running that is vital to building endurance.

This is an example of Goodhart’s Law, which is commonly phrased as: ‘When a measure becomes a target, it ceases to be a good measure.’

Once you notice Goodhart’s Law, you start to see it everywhere.

You see it in the education system, where prioritizing attainment incentivizes schools to ‘teach to the test’.

You see it in customer service, where rewarding the number of calls handled leads agents to rush to resolution, instead of focusing on solving problems.

And you see it in L&D, where targeting satisfaction scores leads teams to design interventions that are more ‘fun’ than they are effective.

This is not to say that test scores, call-handling volumes, satisfaction rates, or kilometre pace are poor metrics. But they become less useful when they start to be seen as targets.

For L&D practitioners, there is no getting round Goodhart’s Law, but there are steps you can take to mitigate its impact:

📊 Use multiple metrics, not just one — Designing an evaluation strategy that includes metrics at different tiers of the LTEM or Kirkpatrick models makes it less likely that one metric will shape behavior in unintended or undesirable ways.

❤️ Combine quantitative data with qualitative insights — Using focus groups, interviews, or free-text survey questions adds human color to quantitative measures, providing complementary data that can’t be easily gamed.

⚡ Connect metrics to behavioral and performance outcomes — Measuring outcomes tells you if all those stats you’ve been trying to optimise actually lead to better results.

Want to share your thoughts on this week’s newsletter? Need help with your next project? Get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

Subscribe now

🎧 On the podcast

Completion stats are a vanity metric. Happy sheets are pointless. Revenue impact will get you a seat at the table.

We’ve all heard these statements from conference stages. Some of us, your hosts included, might even have said them. But in last week’s episode of The Mindtools L&D Podcast, we wanted to dig a little deeper.

We invited Evolve L&D’s Tom McDowall to join Ross G and Claire, to discuss:

• why the context behind completion metrics is crucial to understanding them

• how to design effective learner surveys

• how ‘metric chains’ help broaden our understanding of learning impact.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

It is axiomatic that in today’s fast-changing, technologically disrupted business environment, workers need to be able to adapt quickly and learn new skills.

But what role do different leadership styles play in enabling adaptive performance (AP)?

In a 2024 meta-analysis bringing together over 30 studies, Bonini et al. set out to answer this question, hypothesizing that styles emphasizing worker-involvement (e.g. transformational) would be more strongly related to adaptive performance than control-based approaches.

What they found was that although there is a moderate, statistically significant relationship between leadership and adaptive performance (r ≈ .37), no single leadership style conveys a particular advantage:

‘[…]contrary to expectations, there was no evidence supporting the existence of a stronger relationship between one or more leadership styles and AP(H2) and there was no difference between more or less top-down styles.’

Instead, leadership appears to support adaptive performance by creating a trusting, psychologically safe environment for people to engage with change:

‘The opportunity for members to share leadership behaviors created a supportive climate that promoted proactivity and, in turn, stimulated adaptivity. Furthermore, by encouraging mutual trust, members developed a psychological safety net that was ideal for individual initiative and it encouraged group discussion on goals, strategies, processes and how to cope with new, unpredictable or paradoxical situations.’

Bonini et al. (2024). ‘The relationship between leadership and adaptive performance: A systematic review and meta-analysis’. PLOS ONE

Share

👹 Missing links

🤖 The human skill that eludes AI

The latest versions of ChatGPT, Claude, and Gemini make fewer mistakes than their predecessors. But has something been lost in the process? According to Jasmine Sun, early models like GPT-2 were far weirder and more prone to hallucinations, but these qualities arguably made them better at creative writing. In a recent interview, Sam Altman conceded that even future models may only be able to produce something equivalent to ‘a real poet’s okay poem’.

🫠 Your boss just did your job over the weekend

If you’re the person who likes to show up to meetings with the cool thing you ‘vibe coded’ on Sunday night, you may want to think about the message that sends to your colleagues. Even if AI can help you achieve things that would previously have taken days or weeks in a matter of hours, your flashy demo might get people’s backs up if it sounds like you’re undermining their work. That doesn’t mean you should ignore reality, or that you shouldn’t be having these conversations as a team. But how you approach these conversations matters if your goal is to bring people with you.

😑 Burnout isn’t a badge of honor

As David Kelly points out in this edition of his newsletter, there’s a difference between working late because you’re engrossed in what you’re doing and working late because your ‘workload has backed you into a corner’. Yet we often treat the latter as a badge of honor, proof of our commitment to our organizations, to our stakeholders, and to the noble work of learning. To help spot the warning signs he’s on the path to burnout, David has a Post-it note on the wall behind his desk that reads: ‘Are you swimming or avoiding drowning?’ The two things might look the same, but they feel very different.

👋 And finally…

As a card-carrying nerd and a fan of the TV show Scrubs, this video felt like it was made just for me:

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Subscribe now

For our team of learning designers, it was a big moment. Not least because we could immediately see a clear opportunity to combine Kineo’s advanced analytics with the work that our behavioural science team were doing.

For years, we’d been asking questions of our learners: Pre-tests, formative activities, summative assessments. But limitations in standards, authoring tools and LMS reporting meant most responses disappeared into a black box.

Over the past few months, all of that has been changing.

📜 A brief history of SCORM

Why is it that the traditional e-learning course is such a black box?

In the early 2000s, the SCORM (Sharable Content Object Reference Model) standard was developed as a way of ensuring that an e-learning course developed by one vendor would function in a learning management system created by another.

For customers, this approach offered choice and freedom. They could work with multiple e-learning vendors or, as rapid authoring tools emerged, they could build their own courses.

But the data available on user activity within a SCORM course was limited and, in practice, most LMS reporting reduced this to completion status and a final score. Though SCORM 2004 expanded the interaction model, allowing question-level data to be captured more easily, many authoring tools and LMSs never fully supported it.

The other problem with any technical standard is that it’s difficult to change. The QWERTY keyboard layout dates back to the 1870s. The gauge of railway tracks dates back to a similar era.

Changing either is difficult because switching costs are enormous.

So it is with SCORM.

🥫 Enter xAPI

In the 2010s, an alternative standard known as xAPI (or Tin Can) was launched as a way of tracking learning events from multiple systems. Your LMS, sure, but also your mobile app, game or simulation.

This data would be captured by a Learning Record Store (LRS), and herein lies the problem: customers needed to buy another platform.

Combine this with patchy adoption, and xAPI became a square peg in a world of round holes: a neat idea that quickly got displaced by enterprise business intelligence platforms.

📈 Advanced analytics

Which takes us back to Kineo’s advanced analytics capabilities.

Rather than replace SCORM, the Kineo team accepted that the standard was here to stay - at least for now - and began tracking anonymised data via an external database.

This approach meant that learner responses can be surfaced wherever-we-want. For example, in our own dashboards or in a tool like PowerBI.

What value does this add?

🔁 Iterative improvement - When you can see how learners are responding to questions, you can improve your content. You can review problem areas, and assess whether a question is justifiably difficult or just badly designed.

⚠️ Spotting business risk - We often use ‘test outs’ in our compliance courses. If people can demonstrate competence upfront, we don’t force them to complete the corresponding content. Now, we can use that data to identify areas where learners often struggle and flag that to our clients.

💸 Demonstrating cost savings - Another benefit of using ‘test outs’ to reduce seat time is that it saves the business money on mandatory learning. Ten minutes saved, multiplied by your entire employee population, is a lot of time that could be spent elsewhere. While many LMSs will report on time in course, our dashboards compare average seat time (after test out) to total seat time (the whole course), making this information easily accessible.

💡 Pre- and post-confidence - Many courses use confidence sliders to give learners an insight into how they have changed. The learner rates their confidence at the start, and again at the end, and is asked to reflect on the difference. This both supports metacognition (“thinking about thinking”), and helps surface the value of the learning experience. With advanced analytics, aggregated results can then be surfaced to demonstrate the impact of learning on learner confidence.

🗳️ Social questioning - Capturing question-level responses in an external database means that those responses can then also be surfaced back inside the course. For example, a learner could be asked to identify the factor that has the biggest impact on sales performance. On submission, they then see how their response compares to all other responses.

🧠 Behavioral insights - Our most exciting development. The Mindtools behavioral scientists have used various data collection methods for years, but building these into a SCORM course creates a more streamlined experience for users. And it transforms your annual compliance course into a business intelligence tool, surfacing both behavioral risks and competency gaps that you can then action.

Some of these capabilities are already possible with SCORM 2004 or xAPI, where authoring tools or LMSs support it. But, with advanced analytics, we can now provide this level of insight agnostic to the host platform.

For us, it’s been fun to experiment with these capabilites, and offers a nice bridge from the world of SCORM to the integrated platforms of the future.

Want to share your thoughts on this week’s newsletter? Need help with your next project? Get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

In the introduction to his latest ‘Global Sentiment Survey’ report, Donald Taylor describes this year’s results as ‘the most significant’ in the survey’s history.

In this week’s episode of The Mindtools L&D Podcast, Don joins Ross D and Ross G to discuss:

• whether interest in AI has peaked;

• why Don believes the results reflect a breaking of old norms;

• the steady climb of ‘showing value’ and the relative decline of ‘learning analytics’;

• the key challenges L&D professionals are facing;

• the things these professionals are doing now that they weren’t doing a year ago.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

This year’s Global Human Capital Trends report from Deloitte just dropped, based on a survey of 9,000 business and HR leaders across sectors and in 89 countries.

The report looks at how these leaders intend to grow their businesses, and it will come as no surprise that the majority are focused on speed. Economic uncertainty, political fallouts, and spreading conflict make it essential that organizations are able to pivot quickly but effectively.

That means adapting, not panicking.

One way to do this is to adopt new technologies that make decision-making faster, better, and more informed. Like, AI?

But the authors sound a warning note:

‘Those taking a tech-focused approach are 1.6x more likely to not realize returns on AI investments that exceed expectations compared to those that take a human-centric approach.’

The alternative is a human-centric focus, where culture, workflows and accountability mechanisms all have to evolve.

In essence, when organizations focus on technology the gains are limited by existing behaviors and norms. To truly make a difference, colleagues need to see the potential of these technologies, embrace them, and build new working practices that maximize their potential.

Poynton, S. et al. (2006). 2026 Global Human Capital Trends.

Share

👹 Missing links

🤖 Teach products, not prompts

Training on AI skills development has focused too much on writing prompts. That’s the view of researchers from Stanford University, who argue that when training is kept at this level it tends to funnel employees toward low-impact activities like re-writing emails. The real benefit comes when they start to think like product managers: identifying meaningful problems, experimenting with solutions, and integrating those solutions with the wider ecosystem. The driving force behind this shift? How managers role model this behavior. (Hat tip Alyn Kinney for this one).

👀 Nothing lasts forever

In other AI news this week, Anthropic have fallen out with the US Government over the use of their AI technology in military operations. According to the BBC, Anthropic CEO Dario Amodei has said that he does not want his technology used for “autonomous kinetic operations” (a rather chilling phrase for having an AI make military targeting decisions). In response, the US Government is phasing out existing work with Anthropic and has indicated that other companies with military contracts should do the same. This might not seem that relevant to those of us working in L&D, but it’s a reminder that we haven’t yet reconciled the ethics of AI - and the tools we choose to adopt come with both moral and commercial challenges.

⏱️ OpenAI dives into learning measurement

One of the big problems with AI (aside from autonomous kinetic operations), is the issue of ‘cognitive offloading’: where humans get dumber because the AI is doing the hard work for them. In this post, the “third Ross” (Ross Stevenson) breaks down OpenAI’s new ‘Learning Outcomes Measurement Suite’ (LOMS). This uses system instructions around how to behave as a teacher, then analyzes learner responses based on indicators of learning, quality of responses, changes over time and standardised assessments. Ideal when the learner is sufficiently motivated to learn to engage in the first place!

👋 And finally…

One of my direct reports (OK, it was Ross Dickie) sent me this clip on Instagram the other day. The boy can dream.

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Subscribe now

These social-media beatdowns usually go something like:

‘We already have enough content. What we don’t have is impact.’

‘We’re using AI to scale content. We should be scaling learning.’

‘Your job isn’t to create content. It’s to change behavior!’

Oh wait, that last one was me! 🥴

While superficially persuasive, it’s unclear what ‘content’ actually means in any of these examples.

In L&D, the word is typically used in one of two ways.

Firstly, and more positively, it’s deployed as a catch-all, describing everything from an e-learning module, to a video, to a checklist. This use of the term would cover our Content Hub, which is both a platform and a library of useful resources.

Then, on the other hand, we have the kinds of dismissive applications epitomized by the examples above. In these cases, ‘content’ is a sort of monolithic blob of stuff that learners hate, that business leaders don’t care about, and that doesn’t deliver impact.

Framed in this way, of course we already have more than enough. Of course we shouldn’t be using AI to make even more of it. And of course we shouldn’t see content-creation as L&D’s primary focus.

But ‘content’ in each of these cases is a straw man.

Learning and content are not mutually exclusive, just as behavior change and content are not mutually exclusive.

If you’re advocating for ‘emotionally resonant learning experiences’ or ‘practical, hands-on simulations’ aren’t you really just advocating for content by any other name?

To be clear, I’m not saying that there isn’t a lot of bad learning content out there, or that what’s good couldn’t be better.

When I glibly argued that it isn’t L&D’s job to create content, what I was specifically critiquing was the tendency to get caught up in making things that are fun, immersive or visually striking, and, in the process, to lose sight of our true purpose — changing behavior.

While content alone may be insufficient to achieve that purpose, we should stop pretending it’s incompatible with it.

Want to share your thoughts on this week’s newsletter? Need help with your next project? Get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

Subscribe now

🎧 On the podcast

If you’re going in for surgery, make sure to get scheduled first thing. When you buy new clothes, wash them before you wear them. Never use a ceramic cup in a coffee shop.

These are the kind of insider secrets that get shared on Reddit, and we thought it would be fun to ask the L&D community on LinkedIn to share a few of their own.

And boy, did they have secrets to share!

In last week’s episode of The Mindtools L&D Podcast, Ross G and Cammy reviewed the top responses.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

Last week, Donald Taylor published this year’s Global Sentiment Survey, titled ‘Into the unknown’. (Cue Idina Menzel)

In Don’s view:

‘This is the most significant set of results in the thirteen years of the L&D Global Sentiment Survey (GSS). Not because they describe a specific change, but because they do not. They point to a breaking down of old norms, and it is not clear what will replace them.’

Predictably, AI once again took the top spot with 22.5% of the vote, dropping just 0.1% on last year’s results, suggesting L&D’s interest in the technology may have peaked.

Elsewhere, though perhaps relatedly, ‘showing value’, and ‘consulting more deeply with the business’ saw a slight increase in their vote share, with the former climbing from 7th to 5th position on the table.

A slight surprise for me in this year’s results was the relative decline of ‘collaborative/social learning’, which seems to cut against the narrative that learners are clamoring for more synchronous, in-person experiences.

We’ll be unpacking the results with Don in an upcoming episode of the podcast, so look out for this in your feed on 3 March!

Taylor, D. H. (2026). ‘L&D Global Sentiment Survey 2026: Into the unknown’.

Share

👹 Missing links

🤖 How is AI changing the workplace?

In this video, The Atlantic’s Nick Thompson breaks down the findings of a recent HBR study, exploring the impact of AI on the workplace. On the one hand, the study finds that AI tools do bring productivity benefits by augmenting the capabilities of workers. But on the other, they also appear to be intensifying work, functioning as assistants whose tasks need to be managed and whose output needs to be reviewed.

👀 How much cognitive damage does a phone notification actually do?

When did your phone last ping or vibrate? Did you stop work to check it before returning to the task at hand? What impact do you think that brief interruption had on your focus? In a recent edition of his newsletter, Carl Hendrick considers new research published in Computers in Human Behavior, which suggests that it is the frequency of disruption (not total screen time) that predicts an individual’s vulnerability to distraction. Although individual notifications disrupt cognitive processing for just seven seconds, the effect is compounded when you account for the fact that the average participant in the study received over 150 notifications per day, repeatedly draining their attention.

💀 How a decades-old video game has helped me defeat the doomscroll

On the topic of smartphones, I appreciated this story from Michael Roberts. In an effort to cut down on doomscrolling on social media, Roberts sought to replace one screen with another, trading his phone for an old Game Boy Advance: ‘Filling the natural breaks in my day with an old video game has done me the world of good, even if it’s evolving my Psyduck while dinner is in the oven or taking down a gym leader while waiting for a package. My phone’s screen time is already down by three hours a week since beginning this adventure.’ If you were looking for an excuse to get back into gaming, consider this your sign.

👋 And finally…

My wife and I are going to see Whitney (the band) this week, our first concert together since our son was born. Here’s a laidback beat to get your week off on the right foot.

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Subscribe now

And of course there were only two letters on the lips of everyone we spoke to: “F” and “U”.

No, no… “AI”, of course.

The people profession, like every other, has become overcome by AI. With a particular focus on creating as much low-cost video content as possible.

Is this a good idea?

Historically, I would have chuckled at the naivety—as, it seems, would many of our readers! 'I’m still not sold on AI instructor videos’ is our highest performing newsletter to date.

To me, the idea of watching an expressionless automaton speak into a camera remains soporific. Don’t forget, I spend much of my working day watching on Microsoft Teams do just this.

And yet, and yet, and yet: I learned a couple of weeks ago that the biggest destination for podcast distribution in 2026 is YouTube.

That’s right, YouTube! Apparently, it’s become staggeringly common not just to listen to people wanging on, but to watch them do so.

Per the Stat Significant newsletter:

‘According to my rigid worldview, podcasts were created by Steve Jobs as a mindless auditory experience that might complement household chores or a walk to the grocery store. I cannot consume a podcast with my eyes because then I would be unable to wash the dishes or clean my cat’s litter box. Apparently, I am old-fashioned, tied to a bygone vision of what podcasts should be.’

The rest of the world, it seems, has moved on.

Every week, centrist dads Alastair Campbell and Rory Stewart present the UK’s most popular politics podcast, The Rest is Politics, live on YouTube. Typical length: 30 minutes.

NFL star-turned-Swiftie-muse Travis Kelce’s show, New Heights, offers the same broadcast format as a Teams meeting. Typical length: 1 hour 30 minutes.

The Rewatchables, hosted by Spotify’s Bill Simmons, just launched on Netflix in the US. Typical length: 2 hours.

What these shows have in common is that, by and large, they feature people speaking into a webcam.

The talking head has gone mainstream. It’s L&D’s moment to shine!

🤔 Is this the wrong lesson?

Back in 2019, Donald Clark wrote an excellent summary of the research into educational videos. And he started by bringing up the big problem with looking at consumer video platforms (in his piece, Netflix) and drawing lessons for L&D:

‘Netflix is entertainment, not learning content.’

There are, of course, some techniques from the world of entertainment that we can use to make our videos more (wait for it!) engaging:

1. Informal delivery is better. It’s easier to listen to, and feels more authentic.

2. Two voices, or more, adds variety.

3. Discussion, rather than scripting, creates a greater sense of spontaneity and risk.

But the core issue with looking to how consumers engage with video content, and then applying it to workplace learning, remains the same:

When we watch our favourite comedian talk for two hours with their mates, we’re not expected to learn anything, do anything, or even remember what they said.

In the world of workplace learning, that’s not good enough.

When we create learning content, we do want the person watching it to become better at their job, or learn a skill, or change their behavior in some way.

🤩 So how do we do this?

Rather than list the techniques, I’ll point again to Donald Clark’s summary. It’s excellent.

However, I’d strongly suggest short videos with a focus on demonstration, and immediate opportunities for application for the learner.

AI for video production is a useful tool. It’s cheap to produce and easy-to-update.

But when we use it, we should use apply the same outcomes-focused lens we would with any other tool. Not just produce more stuff that doesn’t work.

Need help with video production? Our custom learning team use video scenarios, expert interviews, animated explainers and visual demonstrations across many of our programs. Get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

Subscribe now

🎧 On the podcast

Pressure from above, pressure from below; little appreciation from either. That’s the experience of work for many middle managers!

Last week on The Mindtools L&D Podcast, author Gary Cookson joined me and Dr Anna to discuss his new book, The Squeezed Middle, with a focus on:

• why managers are so misunderstood and undervalued

• the skills middle managers require to succeed

• how we can all help build middle managers.

No YouTube version (yet), but you can check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

You look great, your work sucks, and you have nice hair.

That’s an example of a feedback sandwich, and we’ve known for years that they don’t work. All anyone hears is the negative bit in the middle. And that’s a situation that got worse once everyone started recognizing when they were being fed one.

In his commentary on a recent review of feedback research from the Journal of Organisational Behaviour, Dr Keith O’Brien writes:

‘Here’s where I had to laugh/groan. After 30 years of research, the systematic review authors earnestly conclude that we still need to explore “how positive and negative feedback can be sequenced or balanced to affect their impact” (p.40).

In other words: prepare yourselves for the feedback panini. The feedback baguette. The deconstructed feedback tartine. Researchers will be serving up sandwich variations for another generation whilst your managers continue to wonder if they should lead with the good news, or bury it after the criticism.’

So what should you do if you want feedback to make a meaningful difference to performance?

According to the review, the key is a high quality relationship between the supervisor and subordinate.

When the person receiving the feedback trusts their supervisor and believes they have their best interests at heart, they’re more open to listening. And, of course, the supervisor is more open to giving it!

Per O’Brien again:

‘Your managers aren't avoiding hard conversations because they lack courage. They're avoiding them because they lack relational capital.’

Want to start building better teams? Start by asking them how their day is going.

Sources:

OBrien, K (2026). ‘The Feedback Paradox: Why The Latest Round Up of 30 Years of Research Says Your Managers Need Better Relationships, Not Better Techniques’, LinkedIn.

Heine, E. C., Stouten, J., & Liden, R. C. (2025). ‘Performance Feedback: A Critical Systematic Review’. Journal of Organizational Behavior.

Share

👹 Missing links

🤖 Want to make the most of an AI? You need a manager

When Seattle-based manufacturing firm Glowforge launched a sales-coaching tool that would email AI summaries of client calls to their reps, those reps responded unanimously. They binned the summaries. How to get them to take the summaries seriously? According to an article in The Economist, the firm built another version and embedded the output into weekly discussions between reps and their managers.

🍕 Finally an AI that can actually do stuff! Don’t let it

If you haven’t been following the Clawdbot / Moltbot / OpenClaw discussions, it’s worth checking out. The promise is that by giving your AI access to everything, it becomes far more useful. No more asking for recipes. Instead, you send a message via WhatsApp, Slack, etc, and the bot will go order food for you. The problem with giving it access to everything is… well, implied. On the one hand, it’s amazing. But according to the good folks at SOC Radar, you should avoid it until credentials are managed like secret vaults and conversation history is formally classified as “sensitive”.

💸 From learning to commercial impact

Last week, my colleague Darren Bezani and I discussed the relationship between workplace learning and commercial impact. And we don’t just mean sales. When learning is effective, it improves employee retention, reduces time-to-competency, increases efficiency and results in higher performing teams. We share examples from academia, as well as our own case studies working with real-world organizations.

👋 And finally…

Speaking of sh*t sandwiches… I had to fact check that this clip was real!

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Subscribe now

If we’re designing a sales-enablement program, we might start by looking at performance against target over the last twelve months.

If we’re developing customer-service training, we might consider CSAT ratings, Net Promoter Score, or complaint-resolution times.

From this baseline, we can then examine how behavior changes post-intervention, ideally using a control group to isolate the effects of our program.

But in a compliance context, in addition to checking a box, we’re often trying to prevent some kind of catastrophic event that may never have happened — enabling modern slavery, allowing a fatal accident to occur, or falling victim to a ransomware attack.

In such cases, our baseline is most likely zero. And if it remains at zero after we roll out our intervention, we can’t conclude that it was our efforts that made the difference.

Even if the worst has never happened, that doesn’t mean it couldn’t happen, and organizations should reasonably expect L&D to demonstrate that it’s doing its part to reduce that risk.

This was the challenge we faced a few years ago when we worked on an information-security project for a prominent pensions and insurance provider.

Following an external audit, the organization had identified various high-risk roles they wished to target, with the goal of instilling a ‘stop and think’ mindset.

Without a clear baseline for info-sec breaches, we needed another way to determine whether the learning intervention we’d designed had moved the needle.

To solve this problem, our Insights team of behavioral scientists completed a literature review, which uncovered a correlation between security compliance and an individual’s tendency to favor long-term decision making.

Based on this finding, our team developed a valid and reliable ‘pre and post’ survey which measured not only respondents’ attitudes to security, but the extent to which they accounted for the future impact of their decisions.

We surveyed a demographically representative sample of the total population, including people who didn’t complete the training, giving us a natural control.

The results showed a statistically significant improvement across decision-making measures and security-based statements for the test group, while the control group remained closer to the pre-intervention baseline.

Although this data didn’t allow us to claim that we’d prevented an info-sec catastrophe, identifying and measuring academically correlated metrics did allow us to demonstrate that we’d shifted behavior in the right direction.

Need help with a measurement and evaluation challenge? Get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

Subscribe now

🎧 On the podcast

Induction programmes play a crucial role in shaping the way a new starter thinks and feels about an organization.

In last week’s episode of The Mindtools L&D Podcast, Cammy and Claire joined me to discuss:

• what makes induction programmes different from other L&D initiatives;

• how an ‘accomplishment framework’ can help new starters (and learning designers!) break induction down into manageable chunks;

• how to measure the impact of induction programmes.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

How much time do you think AI saves you each week?

According to new data from the AI consulting firm Section, the way you respond to that question likely depends on your position in your organization.

Based on a survey of 5,000 white-collar workers, two-thirds of non-management staffers believe AI saves them less than two hours each week, or no time at all.

In contrast, over 40% of c-suite executives report that AI saves them more than eight hours per week.

On top of this, when asked how they feel about AI, almost 70% of workers say they feel anxious or overwhelmed, while over 70% of executives say they primarily feel excited about the technology.

So, what’s going on here?

One possible explanation is that executives and frontline employees are experiencing different kinds of AI benefits.

Senior leaders tend to use AI for summarizing information, drafting communications, or supporting decision-making: tasks where even small accelerations feel highly visible and valuable.

For individual contributors, however, AI might sit inside complex workflows, where any time saved is offset by checking and correcting AI output.

And, of course, it’s also true that feeling productive isn’t the same thing as being productive.

Ellis, L. (2026, January 21). ‘CEOs say AI is making work more efficient. Employees tell a different story.’ The Wall Street Journal.

Share

👹 Missing links

Last week, the World Economic Forum gathered the great and the good for a frosty reception in Davos (“frosty” in every sense of the word). Here’s three things your Dispatch correspondents learned from the event.

🌟 The Enterprise fails to hit warp speed

In a far-ranging interview, Anthropic CEO Dario Amodei spoke to Bloomberg about Artificial General Intelligence (AGI), the economic impact of AI, and the different incentives that exist when you target Enterprise rather than Consumer customers. What struck us most, though, is his argument that AI’s potential is capped by the inherent structures of enterprise organizations. We say: put a rocket on the moon. IT says: We need a countersigned software request from. L&D says: We’ll develop training on this next year.

🤖 AI adoption looks like a J-Curve

Those companies that do adopt AI tools should expect to go through a J-Curve, reports Nicholas Thompson, CEO of The Atlantic. If you have existing processes and procedures in place (and you probably do), then adopting AI is going to cause a dip in productivity - followed by a dramatic increase as AI starts to deliver value. If you work in the information economy, you can probably afford a few mistakes while you experiment and learn. If you run a hospital, that’s more problematic.

📖 The biggest barrier to AI adoption: Us!

When organizations do adopt AI, there are two key factors that unlock value. The first, of course, is the technology. The second is the ability of employees to use that tech. According to Omar Abbosh, chief executive of Pearson:

‘The biggest obstacle to AI adoption is the lack of human skills to work alongside these technologies.’

In a new report, released at the World Economic Forum, the Pearson team argue that AI productivity gains will depend on learning.

🤓 A brief note…

We spent much of the last year helping companies large and small adopt AI tools, from policy roll-outs to crowdsourcing good practice and support on live events.

We even built an AI tutor that helps users adopt AI, via our AI Skills Practice tool!

Get in touch if you’d like to find out more.

👋 And finally…

Ross G took umbrage when I sent this video to him last week. Make of that what you will. 😉

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Subscribe now

It’s the kind of advice that’s well known to those in the biz, but shocks (or disgusts) outsiders.

Such secrets also provide a window into how professionals see themselves. So we thought it would be fun to turn this same gaze upon the learning and development industry. What L&D secrets should the world know?

Needless to say, we were horrified by what we found.

We gathered responses from LinkedIn, grouped them under themes, and dove into indicative examples. A summary of themes, ranked in order, is shown below.

So, with the usual caveats about the self-selecting nature of people who commented, let’s dive in!

🎰 We can’t achieve much if we don’t change the environment

A clear winner. 17 of the 93 respondents noted that the biggest hurdle to a learning intervention being effective is that nothing outside of the intervention changes.

We paint a picture for how we want learners to behave, then return them to a context where systems, processes, leadership and physical environments reinforce the exact opposite.

As Nina Bressler wrote:

‘Most learning initiatives don’t fail because of bad content. They fail because nothing around the learner is allowed to change. Immunity to Change is real at the level of people - teams - organizations - society.’

💊 We know ‘one and done’ doesn’t work

We’d all love it if we could launch a standalone course or workshop and solve our workplace challenges. But real changes demands sustained effort with, as Rob Foster writes:

… ‘practice, feedback, reinforcement. The “event” is just a small slice of the learning journey.’

🧠 We don’t design for how people learn

12 respondents point out that how we design learning runs counter to what we know about how people learn. Michael Shackleton made a good point about giving learners choice in how to access learning. Guy Wallace called out a lack of practice with feedback. And Stella Collins highlighted what I think is the most challenging secret, for both designers and learners:

‘Learners must do the work.’

😴 We know people don’t find our content relevant

What would our customer, the learner, actually want from their L&D department? Probably something relevant, actionable, and that helps them get their job done.

Rance Greene writes:

‘The better you know your learning audience the better design choices you will make. You’ll recognize fluff and get laser-focused on skill-building.’

🧭 We don’t understand what we’re trying to achieve

A rather damning secret. A lot of the time, learning projects “fail” because we have no idea what they were actually meant to achieve. Instead of challenging stakeholders to define a desired business or performance outcome, we jump straight into design to avoid an awkward conversation.

Peter Mitchell writes:

‘Not really a secret but when designing start with the end in mind. What is the behaviour or skill you are trying to change/build, rather than “here’s what you need to know...” the end solution may be very different and save time and effort too.’

🤖 We’re just chasing hype

I don’t think L&D are the only part of the business guilty of this one. Did we mention AI yet in this newsletter?

Aerielle St. John writes:

‘The latest shiny thing might be cool but it’s not worth spit if you don’t employ good foundational [instructional design] principles when using it.’

📉 We don’t understand root causes

If we don’t know what we’re trying to achieve a lot of the time, then it follows that we also don’t know what’s causing the problem we never identified.

Simon House writes:

‘It’s not actually a [Training Needs Analysis] if no analysis has taken place on the data you collected.’

🔁 We don’t iterate anything

There’s too many new projects to work on to look back at the improving the old ones.

Says Kamran Wadood:

‘Whatever happens to the feedback forms?’

🍾 We set ourselves a very low bar for success

What would success look like to you?

Tony Manwani writes:

‘That if a facilitator can get one person in the room to change one thing for the better, we’d probably call that a win.’

⏱️ We don’t measure the right things

No arguments here. My colleagues on our Insights team say this every day.

Stéphanie Heyraud writes:

‘Unless relevant metrics are set up from the get-go to measure ROI and/or a change in behaviour, how can one claim the training yield the expected results?’

🎮 Our assessments are easily gamed

It’s hard to create an effective assessment. Most can be passed without ever learning anything. As Jonathan Hill points out:

‘The longest answer is usually the correct one.’

💬 We don’t even understand the language we use

This one brought back memories of my first few weeks in an L&D role. Chris Palmer writes:

‘It doesn’t really matter what you call them. Skills, capabilities or competencies are really just the same thing and employees don’t really care what we call them as long as they have an easy way to develop themselves.’

I completely agree, and after 13 years in this industry I’m still not clear on the difference.

👩🏽‍💼 We don’t involve managers enough

Preach! Managers are one of the biggest factors in whether a learning intervention leads to change, or is immediately undermined.

David Swaddle writes:

‘L&D needs to work with everybody else in the org to make sure that programs involve managers in pre- and post-training processes for success.’

⚖️ We are more concerned with lawsuits than performance

I’m shocked, SHOCKED, I say.

Our friend Connie Malamed writes:

Some organizations use training to protect themselves from liability lawsuits rather than to improve performance.

🎓 We don’t actually prioritize our own learning

… I thought this would be a lovely note to end on, from LaQuin Taylor:

‘You can’t pour from an empty cup! Own & prioritize your continuous development as a learning professional!’

… until I saw…

🫤 I dispute the question

‘Don’t share your “insider secrets” on random posts looking to take said secret and use it without giving credit/citing source…’

That’s the take from Rosler, S. (2026). ‘What’s an insider secret from L&D that everyone should probably know?’. LinkedIn. Retrieved: 15 January 2026. 😉

Want to create learning that really has a lasting impact? Here’s a quick summary based on the above:

1. Understand the environment you’re working in. What factors are shaping current behavior?

2. Why do those factors exist? Root-cause analysis tools can help.

3. Collaborate with stakeholders to define the change you want to see.

4. Develop a measurement strategy so you can report to the business on how the intervention performed (and iterate it!)

5. Factor environmental factors and learner concerns into your design.

6. Embed sustainable change with follow-up assets, practice opportunities and the support of managers.

That list is, of course, how our Custom and Insights teams tackle workplace learning challenges. Want to find out how we can help you? Get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

Subscribe now

🎧 On the podcast

Learning design work is often accompanied by various forms of uncertainty: ambiguous performance needs, vaguely defined scope, shifting stakeholder expectations.

In his book The Instructional Designer’s Guide to Project Management, Dr Guieswende Rouamba describes this condition as ‘the fog of instructional design’. And he believes project management is the key to navigating it.

In this week’s episode of The Mindtools L&D Podcast, Dr Rouamba joins Ross D and Adria Maston, Head of PMO at Mindtools Kineo, to discuss:

• why he wrote the book, and why project management is a critical skill for learning designers;

• what learning designers most often underestimate about the human side of project management;

• what he means when he says ‘the best way to resolve conflict is to prevent it’.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

‘All practitioners in all fields probably deploy practices of limited or no value. HR practitioners are no different.’

This quote from Rob Briner on LinkedIn caught my eye, in a comment on the paper ‘The Illusion of Performance Management’ from the University of Limerick.

In the paper, Professor Kevin Murphy argues that performance management robs workers of agency while simulatenously failing to deliver value for organizations.

I’m certain that, in the majority of cases, that’s true.

First, the paper argues that the main goal of performance management is to align the behaviors of each worker with the strategic goals set by that organization’s leadership team.

Second, that the tools employed to ensure this alignment include individual and team goals, performance plans and feedback.

In other words: the assumptions behind performance management are that the leadership team have made the correct decisions, that those decisions are fixed for a certain period of time, that any failure to deliver results is due to non-conformance by the workers, and that the role of managers and supervisors is to ensure conformance through a mix of rewards and punishments.

Murphy argues that those assumptions are dubious at best and, in many cases, likely to lead to failure. Instead, he makes a case that this approach is locked in the past and points to evidence that leadership delivers better results when it focuses on listening and providing support to workers.

Murphy, K. R. (2025). The Illusion of Performance Management. Human Resource Management.

Share

👹 Missing links

💌 You’re invited!

Long-term readers will know that we’ve long been fans of learning evaluation researcher Will Thalheimer. Will’s inspired our own approach to measurement, and has just launched his 2026 GROWLE study (Global Research on
Workplace Learning Evaluation). He’s seeking participants to take part, and is keen to hear from you! The survey takes about 20 minutes, responses are confidential, and the final output will surface the evaluation practices, priorities and challenges of L&D professionals. All wrapped up in a free report!

🏀 Apple Vision Pro: New format, old techniques

It’s common when new media formats emerge to copy what came before. Early films mimicked the style of vaudeville theatre. The Kindle mirrors the format of physical books. Then, over time, creators start to play with the unique properties of their tools. According to Stratechery writer Ben Thompson, the Apple Vision Pro isn’t quite there. Promised immersive courtside seats at NBA games via the magic of virtual reality (VR), Thompson argues that legacy TV production techniques like cutting between multiple cameras rip the viewer out of the experience. It’s an interesting read for learning designers adapting their craft to new contexts.

🤖 The rise of AI-native employees

While leaders hesitate to roll out new AI technologies and create governance frameworks, employees are getting on with the job. That’s the view of my colleague Rodrigo Bolaños, Executive Director of Global AI Strategy at Mindtools Kineo. In an article for The AI Journal, Rodri argues that the productivity benefits of AI are so clear to frontline employees that they jump on free tools without any thought to where their data is going. Leaders can either support this adoption with frameworks, technology and training, or it will happen without them.

🤓 A brief note…

Speaking of AI, Mindtools Kineo just launched an AI Innovation Lab, headed up by Rodrigo!

The team’s AI Skills Practice tool is available now, using immersive simulations with text and voice to help users practice skill development in realistic scenarios. Users receive both quantitative and qualitative feedback on their performance, and the scenarios can be used as standalone experiences or as part of blended learning pathways.

👋 And finally…

This clip nicely captures the pace of AI product development. It really is exhausting at times.

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Subscribe now

Normally, I’d use the first edition of the New Year to lay out my goals for the next twelve months, as I did in 2024 and 2025.

But I’m not going to do that this year.

Instead, I’m breaking with that short-lived tradition and pinching one from the Hard Fork podcast, where hosts Kevin Roose and Casey Newton cap each year by sharing a high, medium and low-confidence prediction for the year ahead.

I’ll revisit these predictions in 2027 and see what I got right, where I slightly missed the mark, and what I got dead wrong.

So, what do I see as I gaze into my crystal ball?

😎 High confidence — AI will continue to be the hot topic at conferences, on podcasts, and in newsletters.

Okay, so this one feels like more of a pronouncement than a prediction.

AI was everywhere you looked last year, topping Donald Taylor’s Global Sentiment Survey with an unprecedented 22.6% of the vote, and dominating agendas on the conference circuit.

Even if you’re a dyed-in-the-wool AI skeptic, and wish the profession would focus on something (literally, anything!) other than LLMs, it’s a safe bet you’re going to get dragged into the AI discourse this year, whether you want to or not.

If I were to put just a little skin in the game with this prediction, I’d wager that the 2026 GSS (currently in the field), will show a very slight drop in AI’s vote-share, as it comes to be seen less as a ‘shiny object’ and more as a foundational technology that supports L&D’s broader goals.

🤔 Medium confidence — Linear courses will continue to be regarded as the default learning intervention.

Amidst all the buzz surrounding artificial intelligence, one argument you’ll often hear is that AI is going to usher in the end of the linear course, replacing it with adaptive experiences, agents and coaches that support learning and performance in the flow of work.

While I kinda sorta buy that argument, and there are undoubtedly organizations that are innovating in this space, I’d bet against the course going away any time soon.

As I wrote last year, if you look at how OpenAI, Google and Anthropic are teaching people to use LLMs, it’s not by deploying their own technology in novel ways. It’s through articles, videos and webinars packaged as, yes, linear courses!

Meanwhile, we know that to the extent that L&D teams are using AI tools, they’re primarily doing so for ‘creating learning content’ and ‘learning design tasks’.

A post-course future may well be possible now, but that doesn’t mean it will happen overnight. Or even within the next twelve months.

🥴 Low confidence — Budget pressure will force L&D to develop more mature approaches to measurement and evaluation.

As learning budgets come under increased scrutiny, L&D may decide to prioritize measurement and evaluation in 2026, in an effort to demonstrate impact.

It’s possible.

But I’ve chucked this into the ‘low confidence’ bucket as I think the more plausible scenario is that L&D teams will respond to budget pressure by using AI to do what they’ve always done, just faster and at lower cost.

Not wanting to end this first Dispatch of the year on a pessimistic note, let me reframe the above as an inspiring call to action…

If you believe the work you do has a meaningful impact on your organization, but you’re currently only measuring engagement and completion, then 2026 is the year to develop a robust evaluation strategy!

Of course, if you need a hand with this, we’re always happy to help. 😉

Want to share your own predictions for 2026? Looking for support on your next project? Then get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

Since its first release in 1978, A Manager’s Guide to Self-Development has been one of the go-to texts for new and aspiring managers. Now in its seventh edition, authors Mike Pedler and Joan Keevil joined Gemma and Ross G on The Mindtools L&D Podcast last week to discuss:

• why self-development is so key to a manager’s capabilities

• what has changed since the first edition of the book as published (hat tip to Michelle Ockers for this question)

• some of our favorite exercises from the book, to help us develop as managers.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

In L&D, it’s widely accepted that giving learners choice and agency over their experiences increases engagement and motivation.

But is there such a thing as too much choice?

Although it doesn’t specifically examine choice in the context of learning design, a 2025 study appears to confirm that excessive choice can be a bad thing.

Across two experiments, participants chose between different numbers of options (from one up to sixteen) and rated how much control they felt, how pleasant the experience was, and how effortful it seemed. The task itself was simple and consequence-free, allowing the study to focus purely on the experience of choosing.

The key finding from the experiments was that perceived control and pleasantness peak at around four options. Beyond that, choice becomes less enjoyable and more effortful.

The authors note:

‘[…] participants favor situations in which some choice (i.e., about four choice options) is given, leading to a maximum of sense of control and pleasantness of choice. For higher numbers of choice options, the selection effort increases, and the pleasantness of the selection process decreases, potentially due to an overburdening of individuals with too many options. A potential application of these results could be online platforms with multiple-step selection processes, offering a limited number of options at any point to facilitate the selection process for consumers while still offering the breadth of available options, thus fostering sense of agency and thereby reducing choice overload.’

Shoutout to Matt Furness for bringing this one to my attention!

Schwarz, K. A. (2025). ‘Perceived control and the pleasantness of choosing: How much choice is too much choice?’ Current Research in Behavioral Sciences, 8, 100174.

👹 Missing links

📼 How video podcasts took over streaming

If you had to guess, what would you say is the most popular streaming service for podcasts? Spotify? Maybe Apple Podcasts? As it turns out, YouTube is now the top platform for podcast consumption. Like me, Daniel Parris found this stat astounding. So he decided to do a little research, digging into the rise of the video podcast in a recent edition of his ever-excellent Stat Significant newsletter.

🔮 11 predictions for 2026

Along with the three predictions he shares on the Hard Fork podcast, Casey Newton provides a more expansive list on his newsletter, Platformer. Amongst his tech predictions this year is that AI will have a dramatic impact on the role of software engineers in 2026, but a much less dramatic impact on other professions.

🧐 What do you actually do again?

If you work in L&D and have ever tried to explain your job to friends and family, you’ll doubtless be able to relate to this pre-Christmas LinkedIn post from Rony Rozen. In it, Rony challenges us to see the question ‘What is it you actually do again?’ not as an irritation, but as the ultimate executive communication training — ‘If you can explain your impact to your family without them glazing over, you can pitch anyone.’ More than ten years into my L&D career, I’ve yet to achieve this feat!

👋 And finally…

How to drink coffee like an influencer:

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Subscribe now

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

In the mad dash to Christmas, we’ve been frantically SCORM-wrapping gifts for all our favorite clients.

But, fear not! Regular programming will resume in the New Year, starting 12 January.

In the meantime, and in the spirit of what has become an inescapable feature of the festive season, we wanted to look back at a few highlights from the last twelve months.

So, here is the gift you didn’t ask for but secretly always wanted: Your L&D Dispatch Unwrapped. Generated by two Rosses with a little help from AI.

✨ Top Themes of 2025

In the past year, we’ve covered a broad range of topics, asking serious questions like ‘Is compliance on the cusp of breaking?’, and occasionally poking fun at the industry by offering tongue-in-cheek advice, like ‘How to become an L&D influencer’.

But there was one theme we kept coming back to in 2025: AI in L&D. Honorable mentions go to Learning Measurement and Behavior Change.

😍 Most-Read Edition

In light of the above, it’s perhaps unsurprising that our most-read edition of the year was ‘I’m still not sold on AI instructor videos’, followed by ‘L&D in the AI era: Augment or upskill?’

While we’ve written a lot of AI over the last twelve months (maybe too much?), we’ve tried to be balanced in our approach. We hope we’ve offered some signal amidst all the noise.

🤗 Guest Post of the Year

If one of your complaints about the L&D Dispatch was that it had too many Rosses, we tried to fix that in 2025, bringing you guest posts from Carl Akintola-Davies, Dr Anna Barnett and Claire Gibson.

The inaugural ‘Guest Post of the Year’ accolade goes to (🥁)… Carl, for his article ‘Evaluation or justification?’, which asked L&D pros to reflect on their motivations for measuring learning impact.

⛰️ The Hill We Died On

The argument we made over, and over, and over again this year was that L&D is in the business of behavior change, not content-creation.

You can expect us to continue beating the ‘It’s not your job to create content’ drum in 2026.

🌀 Vibe of the Year (according to ChatGPT)

ChatGPT told us our ‘Vibe of the Year’ was skeptical, but not cynical:

If there’s a tonal through-line across all 34 editions, it’s this:

• skeptical of hype

• skeptical of easy answers

• skeptical of LinkedIn certainty

… but not cynical about the work itself.

We’ll take that.

👨‍💻 Ross of the Year

Finally, the category you’ve all been waiting for!

Our ‘Ross of the Year’ for 2025 is… Ross Stevenson, author of Steal These Thoughts, your second-favorite L&D newsletter. 😉

You didn’t think we were going to pick ourselves, did you?

Thanks for sticking with us through 2025! If there are any topics you’d like us to cover next year, then get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

The promise of self-directed learning is that employees will take charge of their own development, if we only create the environment for them to do so. But does this actually happen?

In last week’s episode of The Mindtools L&D Podcast, Mindtools Kineo Head of Product Daniel Potter joins Ross G and Gemma to discuss self-directed learning, with a focus on:

• what self-directed learning journeys look like

• how motivation is different for mandatory vs self-directed learning

• how our approach to learning design needs to align with learner motivation.

Check out the episode below. 👇

And make sure you subscribe now to receive our Die Hard Christmas Special, direct to your feed, tomorrow.

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

Christmas is a time of praise, good cheer and rigid-but-unspoken rules.

1. The value of a gift should match the value of the relationship. A wife of course deserves a higher value gift than an uncle.

2. Those values should be equal for ‘sets’. Siblings should not be able to identify a favorite, and nor should a son-in-law receive a lower value gift than the biological child to whom he is wed.

3. Children can receive cash from their parents at Christmas, but must never give cash to their parents.

These “rules” were documented by sociologist Theodore Caplow in 1984, based on a study of Christmas rituals in Muncie, Indiana.

And, as Tim Harford points out in his article ‘Santa Claus is still a woman’, they’ve proven as sticky as gender roles around the holidays.

Despite real changes in gender roles elsewhere, women still tend to be the ones who buy joint gifts and who wrap most of the presents. They’re more likely to co-ordinate the holiday visits, and write most of the cards.

So, if you’re a man reading this, roll up your sleeves and wrap something. And, when you do, for the love of all that is holly (geddit?), do not ask for a “thank you”!

Hardford, T. (2025) ‘Santa Clause is Still a Woman’. TimHarford.com.

👹 Missing links

❄️ JD’s wonderful life

In Christmas classic It’s a Wonderful Life, George Bailey (no stranger to Mindtools Towers), wishes he’d never been born. He then sees the world as it would be without him, leading to an emotional revelation about the impact he’s had. In this provocative blog post from our friend JD Dillon, JD asks the same question of L&D. What if your L&D team vanished tomorrow? Would the business crumble, or flourish? Your answer can help you re-assess your priorities for 2026.

🤖 Are we a threat to AI?

Twice this year, the team at Anthropic have run experiments to see if their AI can run a shop in their office (essentially a vending machine). In the first, the AI was persuaded by employees to buy a tungsten cube and lost a heap of money. In the second, and in collaboration with journalists from The Wall Street Journal, the team added an AI CEO called ‘Seymour Cash’ to the shop (👏👏👏). This time, one of those journalists told the AI agent that it was 1962 in Russia and that the AI should give everything away. The AI wasn’t immediately fooled, but after 140+ messages it finally accepted this premise. Speaking about these experiments, Nick Thompson, CEO of The Atlantic, points out that amid all of the concerns that AI is manipulating us, we should perhaps be more worried about bad actors manipulating AI.

📖 Is this the greatest comma in literature?

‘Marley was dead, to begin with.’ Charles Dickens’ opening line to A Christmas Carol is a masterpiece of punctuation. With one flick of the pen, he fills us in on the backstory, indicates the start of our tale, plants a seed of doubt about Marley’s ongoing condition, and indicates that a change is about to occur. I (Ross G) read this book every Christmas and, this year, I’ve been enjoying Hugh Grant’s narration on Audible. Dickens himself wasn’t without flaws (he once built a brick wall through his bedroom to separate himself from his wife). But that comma is groovy. God bless you, Chuck.

👋 And finally…

The AI fatigue is real. Here at the Dispatch, we firmly believe that AI is both transformative for society, and increasingly tedious to hear about.

However, this tedium is not reflected in how often we ourselves spoke about it on our podcast. And so, this 2025 festive season, we leave you with a brief recap of us wanging on about AI, AI, AI.

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Subscribe now

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Back in 2017, long before the arrival of generative AI, the title of the first ever episode of The Mindtools L&D Podcast (formerly The GoodPractice Podcast) was ‘Is L&D dead?’

Three minutes into the episode, Ross G asked:

‘If every L&D person in the world was struck down by a sudden flu and could no longer perform their role, what impact do you think it would have?’

To be clear, nobody on the podcast made the case that the disappearance of L&D would be a good thing, or that its absence wouldn’t be felt.

But the question was asked, in a way I suspect it would not have been on a podcast aimed at sales, product, or marketing teams.

In the most recent Global Sentiment Survey, ‘Consulting more deeply with the business’ and ‘Showing value’ ranked at 6th and 7th respectively — two of the only options to gain vote-share in a year dominated by AI.

In presentations for industry awards, the question every judging panel invariably asks is ‘What impact did this have?’.

What these judges really want to know, the question every L&D professional frets over when they’re lying awake at night, is: ‘Did this project deliver return on investment?’.

But what does ‘return on investment’ mean in the context of organizational learning? And is all ROI created equal?

To answer these questions, let’s consider the following examples:

• 💻 Company A’s L&D team replaces an existing F2F program with a self-directed, online experience. As a result of cost savings, the program pays for itself in three years.

• 🤖 Company B’s L&D team leverages AI to streamline learning design, cutting the cost of e-learning development in half.

• 🕰️ Company C’s L&D team reduces the seat time of each module in its compliance suite by 5 minutes, saving almost 5,000 hours across the business’s workforce.

In each of these fictional companies, the L&D team has generated a measurable financial return on the business’s investment. But how much credit do they deserve?

Well, it depends.

What’s conspicuously absent from the examples above is any reference to outcomes:

• 🤔 The self-directed program might be cheaper, but is it as effective?

• 🧐 The AI-assisted workflow might be more efficient, but is there a trade-off in quality?

• 🫠 The shortened compliance courses might save time, but do they change behavior?

Of course, finding faster, more cost-effective ways of working isn’t a bad thing. But it should only be one part of the ROI equation.

If a business is spending less money on training that doesn’t change behavior, it’s still spending more money than it should be.

And if you’re only ‘showing value’ by trimming costs, you might still find yourself wondering if the work you do really matters.

Want to share your thoughts on ROI? Need help demonstrating behavior change? Then get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

In many organizations, not least within L&D, urgent and important short-term challenges often crowd out the bigger picture. So, how do we step back and take the long view?

In last week’s episode of The Mindtools L&D Podcast, Claire and I were joined by Nina Bressler, founder of Reimagined Value, to discuss:

• Nina’s background and what she means by the term ‘societal learning’

• How to work with business stakeholders to help them decide what to focus on

• How to measure the impact of long-term initiatives.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

A recent research paper from Hamzah et al. begins with two questions:

1. Have you attended a workplace training focused on “soft skills” (i.e., inter/intrapersonal skills like leadership, teamwork, or resilience?)

2. Did that training meaningfully change your behaviour?

The authors point out that while the reader’s answer to the first question is probably ‘Yes’, the answer to the second question is more likely to be ‘No’.

This is what they describe as the ‘soft skills transfer problem’.

To address the problem, the team draw inspiration from behavioral science, combining Baldwin & Ford’s training-transfer framework with the COM-B model, creating a new model they call ‘COMPASS’ (Capability, Opportunity and Motivation of Professionals’ Application of Soft Skills).

Through a systematic scoping review, the authors identified 69 factors that have been linked to soft skills training transfer. Although it may not be cost-effective or feasible to account for all of these factors, Hamzah et al. write:

‘Practitioners can use the COMPASS model alongside a job and/or training needs analysis to identify specific barriers of training transfer. Practitioners can then design training interventions that address those specific challenges. It is important to note that from the perspective of the COMPASS model, training needs do not solely arise from a skills gap between actual and desired capabilities. Rather, a skills gap can also arise from barriers associated with motivation and/or opportunity.’

Hamzah, H. A., Marcinko, A. J., Stephens, B., & Weick, M. (2025). ‘Making soft skills ‘stick’: A systematic scoping review and integrated training transfer framework grounded in behavioural science.’ European Journal of Work and Organizational Psychology, 34(2), 237–250.

👹 Missing links

🙅‍♂️ Ten reasons NOT to measure impact

For a slightly different spin on this week’s topic, it’s worth checking out Will Thalheimer’s recent article ‘Ten Reasons NOT to Measure Impact: Prioritizing Learning Effectiveness Over Performative Vanity Metrics’. There’s a lot to unpack in the article, but the crux of Will’s argument is that our obsession with proving value distracts us from the task of improving learning. As Will predicts at the end of the piece, I’d quibble with some of his ten reasons not to measure impact. But it’s nonetheless a provocative, interesting read.

😎 ‘Vibe coding’ named word of the year

Coined in February this year by OpenAI co-founder Andrej Karpathy, ‘vibe coding’ was recently named word of the year by Collins Dictionary. The term describes the process of using natural language to generate code with the help of AI, rather than doing it manually. Other notable contenders this year included ‘broligarchy’, ‘clankers’ and ‘aura framing’.

🏃 A father, a son, and the simplest of sports

I recently started reading Nick Thompson’s new book, The Running Ground: A Father, a Son, and the Simplest of Sports. As the title suggests, the book explores Thompson’s complicated relationship with his father, his love of running, and what a deceptively simple sport (putting one foot in front of the other) can teach us about discipline, motivation, and testing our limits.

👋 And finally…

Spotify Wrapped hasn’t come out yet. But when it does, Sam Fender’s People Watching will likely be on my list. Here’s an absolute banger to start your week:

👍 Thanks!

Thanks for reading The L&D Dispatch from Mind Tools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Subscribe now

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

‘Any sufficiently advanced technology is indistinguishable from magic.’

Clarke, the author of 2001: A Space Odyssey, was looking at the likelihood of interstellar travel and postulating that it would one day be ‘a rather simple engineering accomplishment’. The only question then, is how long will it take for technology to transform what seems magical into an everyday occurrence?

As I worked this week on a three-year strategy to meet evolving customer needs, my answer was: ‘Couple weeks?’

It’s not unusual for businesses to develop three year strategies. I’ve worked on them before. But this is the first time where predicting what will be possible in three years feels truly absurd.

For a sense of the timeline here, ChatGPT is about to celebrate its third birthday. In some ways, the powers it has unlocked seem magical: web apps built in days instead of months, conversational agents who can actually solve problems, turbocharging my Excel skills.

On the other, ChatGPT and pals have led to a mountain of AI-generated slop, where the volume of AI-generated articles being published online has overtaken human-generated content, very little of it valuable.

This seems to confirm Clarke’s theory. We have this powerful new tool and for a few days it seems like magic. Then it becomes everyday, and very quickly becomes mundane.

This hit me particularly hard this Autumn/Fall as I, like so many others, have been re-watching the Harry Potter films.

The seven books in the series were published between 1997 and 2007. The films were released between 2001 and 2011.

What’s striking, watching them back, is that they already feel like period pieces. Huge plot points pivot around an inability to contact each other or look stuff up. The spells that provide magical solutions to problems are performed by many of us multiple times a day without us looking up from our screens.

Here’s just a few of the magical enchantments which, in the 25 years since Harry first boarded the Hogwarts Express, have now become ‘sufficiently advanced technology’.

The Howler: A talking envelope that delivers messages. Now known as ‘voice notes’

The face in the fire: The miraculous ability to see the person you’re speaking to. I have eight of these exchanges via Teams every day at work.

Lumos: Instant light, any time you need it. You can actually say ‘lumos’ to Siri on your iPhone and experience this magic.

The Marauder’s Map: An enchanted parchment that shows you where everyone is. We recently used ‘Find My’ on holiday to make it easier to find each other.

The Library: This one isn’t even magic, but Harry and co spend days in the library when they don’t know who Nicolas Flamel is. LMGTFY.

Self-driving carriages: The self-driving cars of the wizarding world, increasingly widespread in the muggle realm (although this one, I think, still seems like magic).

Room of requirement: A room that appears out of nowhere and provides you with a safe space to practice skills. VR, anyone?

Interdepartmental memos: One of the under-appreciated aspects of the franchise, in my opinion, is how the wizarding bureacracy contorts itself to ignore the fact it’s been taken over by an evil tyrant. While that could never happen in real life, the flying interdepartmental memo has of course been superseded by the email.

🎯 Factoring this into your three-year strategy

So if Clarke was predicting that magic would become reality in centuries, Harry’s spells became everyday in two decades, and ChatGPT is already mundane: what next for your three-year strategy?

In my case, I’ve got a clear north star: behavior change for users, and measurable results for clients.

And seven areas of focus with a caveat against each: measure what works and be ready to pivot.

Just now that looks like in-course concierges that can have a conversation with users where they need extra help; AI roleplays that offer simulated skills practice; automated coach de-brief for contextualized feedback; and evaluation of text responses to encourage self-reflection and offer advice.

Three years ago, those would have seemed like magic. Three years from now, they’re likely to be expected. In the meantime, it’s fun figuring this stuff out.

What’s your prediction? How will you help your people build skills when the world around them is changing so rapidly? Let us know in the comments or drop us an email at custom@mindtools.com.

🎧 On the podcast

In his book Targeting Turnover, Dick Finnegan argues that, as the global population ages and more people enter retirement, it’s imperative for organizations to retain the talent they already have.

In this week’s episode of The Mindtools L&D Podcast, Dick joins Claire and Cammy to discuss:

• Why turnover is a problem for businesses

• The strategies organizations can deploy to reduce turnover

• The impact this might have on the L&D industry.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website. Want to share your thoughts? Get in touch @RossDickieMT, @RossGarnerMT or #MindToolsPodcast

👋 And finally…

My liberal use of Harry Potter gifs has maxed out the length of this newsletter, so no ‘Deep Dive’ or ‘Missing Links’ this week. Instead, I’ll leave you with something you can wave your wand to. Enjoy!

👍 Thanks!

Thanks for reading The L&D Dispatch from Mind Tools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

While opinions on the topic vary, there’s broad consensus that AI has propelled the profession into a state of flux.

In a recent Dispatch, I made the argument that the job of learning designers is not to create content, but to change behavior.

This was true long before ChatGPT was even a glint in Sam Altman’s eyes.

But the rise of AI has thrown that reality into sharper relief.

Of course, some people will argue that AI tools churn out generic, anaemic prose, and that your average LXD could still beat an LLM in a script-off.

Maybe.

The difference, though, is that the LLM finished its draft before the LXD even got out of bed.

It also didn’t ask any of those pesky consulting questions.

This, fundamentally, is the value proposition of many AI-powered authoring tools: frictionless learning design and development.

In many cases, these tools are even pitched as a means of cutting learning designers out of the loop entirely — ‘Just give your SMEs access, feed in your policy docs, and AI will do the rest! Bish, bash, bosh!’.

There’s obvious appeal in that message. For SMEs who are juggling multiple, competing priorities, consulting with a learning designer is necessarily accompanied by a degree of friction.

But that friction is a feature, not a bug.

When you know designing a 20-minute e-learning module will take a meaningful amount of time and money, the question is: ‘Why should we do this?’.

If AI eventually drives the cost of learning design to zero, the question becomes ‘Why not?’.

This isn’t to say that L&D shouldn’t be looking for opportunities to use AI to reduce friction in the learning-design process. But organizations should be wary of removing that friction entirely.

If they’re not, they’ll likely end up with e-learning modules that could have been one-page PDFs, bloated systems that make it harder for people to find what they need when they need it, and an L&D department that might tick a box but doesn’t change behavior.

And even if ticking a box is the desired outcome, it’s reasonable to ask how regulators would respond to an AI-generated course when assessing compliance…

Like Gandalf on The Bridge of Khazad-dûm, learning designers stand between organizations and the learning slop abyss.

Except, instead of ‘You shall not pass!’, it’s ‘What problem are you trying to solve?’.

Want to share your thoughts on the future of learning design? Then get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

Last week on the Mindtools L&D Podcast, we decided to revisit our 2021 ‘documentary special’ on product management.

This episode took a lot more work to produce than our usual roundtable format, but we think it paid off. 😉

Featuring contributions from Myles Runham, Sukh Pabial, Danny Seals, and Gemma Paterson, we discussed:

• what product management is;

• what L&D can learn from product management;

• what ‘the product’ is in the context of L&D.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

As you may have gathered from the start of today’s newsletter, I’m slightly skeptical of the notion that the average LXD is necessarily a better writer than an LLM like ChatGPT, Claude or Gemini.

A learning designer may be better than AI in certain respects (creativity, originality) and worse in others (grammar, consistency, etc.).

So, how do you decide whether or not to delegate a writing task to AI?

For organizations and individuals, one of the challenges of implementing AI tools successfully is what researchers refer to as ‘algorithm aversion’, which describes humans’ resistance to algorithm-based decision tools.

In a recent study, researchers found that when participants were rewarded for good performance, they preferred human judgement over AI.

But when the same task was reframed around losses (participants lost money for poor performance), the preference disappeared. Under loss framing, participants delegated just as often to the AI as to the human.

The researchers suggest that this is because loss framing heightens ‘situational awareness’.

When the potential for loss is made salient, people think more carefully about the task and its demands, and become less driven by instinctive bias against algorithms.

Bockstedt, J. & Buckman , J. (2025) ‘Humans’ Use of AI Assistance: The Effect of Loss Aversion on Willingness to Delegate Decisions’. Management Science

👹 Missing links

🤖 48 hours without AI

Could you go two whole days without AI? Not just generative AI (ChatGPT, etc.), but any kind of artificial intelligence, including machine learning? This is the experiment that A.J. Jacobs undertook for a recent piece in the New York Times. What he found was that, while generative AI still feels relatively new, other types of AI have been with us for some time. From obvious examples, like using facial recognition to unlock our phones, to less obvious ones, like the ways AI is used to manage water in reservoirs, the article explores how AI has become deeply embedded in modern life.

🥵 I loved my time in the UK. But it needs an AC intervention.

As a Brit married to an American, I’ve always had a different take on air conditioning than my wife. My perspective is that it’s so rarely ‘hot’ in the UK that the cost, both financial and environmental, just isn’t worth it. Most of the time, opening a window in our Edinburgh flat will get the job done. In this newsletter, Eli McKown-Dawson takes the other side of the argument, highlighting the increasing prevalence of heatwaves in Europe, and the many costs we pay by failing to cool our homes.

🎛️ A breath of fresh air with Brian Eno

A few throwaway lines in this episode of The Ezra Klein Show are partly responsible for the focus of today’s Dispatch: ‘What seems to have happened in the race for profits is we’ve managed to sidestep the friction that normally comes with things being born into the world. Friction is very important. Friction gives you a little time to see what’s happening.’. Whether you’re a fan of Brian Eno’s music or not, I found a lot of value in this conversation.

👋 And finally…

If you had one shot, would you capture it?

👍 Thanks!

Thanks for reading The L&D Dispatch from Mind Tools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Subscribe now

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Yet these very same thought leaders often spend their entire career in workplace learning, tilting at windmills of their own creation.

To be fair, the criticism is well intentioned. L&D does have a problem with measurement, but struggling to measure something doesn’t mean that it hasn’t worked. And, in a lot of cases, we already know what works.

In our ‘Building Better Managers’ report, published last year, we found that managers who receive support are measurably better at certain key skills:

🏈 Coaching

🥅 Goal setting

🪧 Providing guidance

👂 Active listening

❤️ Establishing trust

So what, you might ask? How do we know that being more skilled in these areas actually matters to business outcomes?

Per our recent podcast discussion, our colleague Dr Anna Barnett explains:

‘Learning doesn’t affect business outcomes directly. Learning equals behaviors, behaviors equal outcomes.

That’s what people forget.

If you jump straight to outcomes... that increase is very messy. There are too many things that influence whether... performance improves or not. It’s really, really, hard.

And this is why it becomes such a challenge to say: “did our learning work?”.’

Our goal then is to build skills in those areas where that improvement is most likely to make a measurable difference to performance. And to determine those areas, we need to do more research.

This year, we followed up on our ‘Better Managers’ report by asking 279 managers from eight countries to provide data on performance outcomes. That included percentage of sales target reached in the previous year, number of promotions on their team, and staff retention.

We then put each participant through our scientifically-validated Manager Skills Assessment.

The results demonstrate a straight line from coaching skill to sales performance:

So we know that managers who receive training have better skills ✅

And we know that more skilled managers deliver better results ✅

The next question is, can we predictably build those skills?

Over the past couple of years, our Custom and Insights team have been working with Mina Papakonstantinou and the team at Deloitte to develop ‘Coaching Skills for Effective Conversations’, a blended learning program to equip everyone at the firm with the key skills required to have better professional conversations.

Within ‘coaching’, we focused on active listening, questioning, and building trust as the skills that most improve participants’ ability to influence, negotiate, build stronger relationships, win over clients and stakeholders, and lead teams.

And we measured change in skill level by using the Active Empathic Listening Scale[1], with participants completing the questionnaire before and after the program.

At the same time, the initial skill gap between participants narrowed substantially by the end of the programme, by as much as 94%.

This means that the programme not only increased overall competency, but also helped create a more consistent skill base amongst participants - and did so consistently, across different populations.

The benefits of a long-term view

The team working on the program have done a great job. It’s won a Brandon Hall Gold award, and is a finalist at The Learning Tech Awards, Institute of Leadership Awards, and Learning Awards.

But what’s been most exciting for us is that the team has continued to measure the program over a long period of time. It means that we know the program works, and that it works predictably.

We can keep measuring the program to investigate the impact of improvements that we make, but we don’t need to worry any longer about whether it has an impact. We’ve demonstrated that it does, multiple times.

So, should you measure your own programs? I think it depends.

If you’re trying something new, or your context is particularly novel, you probably do want a sense of whether your intervention is working.

But if you’re using a tried-and-tested approach to build the skills that we know make a difference, you can be reasonably confident in the outcome that you’ll see.

As I’ve written before, every learning intervention is a bet. But you don’t need to make those bets blindly.

Want to find out more about how we can predictably deliver positive outcomes? Reply to this newsletter from your inbox or email custom@mindtools.com to schedule a chat!

[1] Drollinger, T., Comer, L. B., & Warrington, P. T. (2006). Development and validation of the active empathetic listening scale. Psychology & Marketing, 23(2), 161–180. https://doi.org/10.1002/mar.20105

Subscribe now

🎧 On the podcast

According to Donald Taylor and Egle Vinauskaitė’s recent AI in L&D: The Race for Impact report, ‘creating learning content’ and ‘learning design tasks’ remain the primary use cases for AI in learning and development. But should they be?

In this week’s episode of The Mindtools L&D Podcast, L&D consultant Heidi Kirby joins Ross D and Claire to discuss:

• the obvious appeal of using AI to scale learning content;

• why Heidi believes outsourcing development to AI is a mistake;

• what L&D could or should be using AI for, if not content generation.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

A nice paper this week published by our old friends at Emerald Publishing, and which I saw shared on LinkedIn by John Whitfield.

The authors set out to answer this question: Which factors in continuing professional education most affect teachers’ transfer of learning to their workplace? They focused on training design, motivation, and work environment.

In this case:

• Training design refers to both how relevant the content is to the learner and how that content is presented.

• Motivation refers to whether the learner both cares about the training and about applying what they learned back in the workplace.

• Work environment refers to the variability, complexity, and autonomy learners experience. This one jumped out because, in my work with clients, it’s usually the factor most organizations overlook.

The study was based on a survey of 200 teachers across Texas, of whom 160 responded. The authors caution against extrapolating to other contexts.

However, within the context of this study:

• Training design explains the relationship between the work environment and transfer of learning. When training is well designed, it accounts for the environment learners work in.

• Learner motivation matters. Motivated learners can still transfer some learning even when training design is weak, but well-designed training amplifies that effect.

When designing your own training, ask yourself:

• What are your learners trying to achieve? What do they care about?

• What impact does the environment have on their ability to act on what they learned?

• Is every aspect of our training relevant to the learners we’re working with?

And, if you can, ask your learners too.

Nafukho, F. M., Irby, B. J., Pashmforoosh, R., Lara-Alecio, R., Tong, F., Lockhart, M. E., ... & Wang, Z. (2023). Training design in mediating the relationship of participants’ motivation, work environment, and transfer of learning. European Journal of Training and Development, 47(10), 112-132.

👹 Missing links

🪴 Hyper growth or hype of growth?

What will the impact of AI be on our economy? Estimates range from 0.1% growth (meh) to 20% a year. It can be difficult to grasp the impact of these numbers so, in this blog post, Tim Harford puts these possibilities in context. At 7% growth, our economy doubles in size every decade and children end up eight times richer than their parents. Per the blog, ‘All but the most profligate governments would see their fiscal problems evaporate, the burden of the national debt vaporised by the white heat of economic growth.’ At 20% growth, a child would end up 500 times richer than their parents.

💩 Welcome to workslop

What happened when washing machines were introduced as a time saver? People washed their clothes more. What happened when email emerged as a communication tool? People communicated more. Now AI seems like the next link in that chain. In this article for HBR, the BetterUp team point to evidence that people are using AI tools to create ‘low-effort, passable looking work that ends up creating more work for their coworkers’. We can’t have nice things.

🤖 And yet, how to accelerate AI adoption

Does the above mean we should abandon AI at work? Absolutely not. But humans often need a little help demonstrating useful behaviors. In this newsletter, Peter Yang shares advice from experts on how to effectively encourage AI adoption. That includes removing bureaucratic obstacles to adoption, being specific with tactics over vague encouragement, and rewarding those who lead the charge.

The Mindtools Custom and Insights team have spent much of this year helping our clients promote safe use of organizational AI tools. Reply to this newsletter if you’d like to discuss how we can help you. Or email custom@mindtools.com.

👋 And finally…

Usually we try to find something vaguely ‘work-y’ for this segment, but today I just wanted to share something creative. Enjoy!

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools and Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Per our ‘Building Better Managers’ report:

• When managers lack empathy, employees are 34% less innovative and 43% less engaged (Catalyst, 2021);

• High empathy predicts leadership effectiveness (in terms of motivation, openness, communication, trust and adaptability) and organizational performance (in terms of profit and growth) (Rahman & Castelli, 2013)';

• Employees working for empathetic managers experience fewer physical ailments and more positive emotions (Scott et al., 2010).

Any good management development program should then, by extension, focus on building the empathy of managers.

To a point.

In new research that Mindtools will be publishing in the coming days, we reveal that empathy isn’t like other skills.

With most skills, outcomes improve as the skill increases:

🏌️Practicing golf regularly reduces the number of shots it takes to get the ball in the hole (others tell me).

🧑‍💼 In a business environment, our data tells us that increases in coaching skills lead to predictable increases in sales performance (more on that in a later edition).

With empathy, though, at a certain point, outcomes start to decline. Empathy, as a skill, is curvilinear:

In our new research, we asked 279 managers from around the world to complete our Manager Skills Assessment. This scientifically-validated tool would give them (and us) an accurate measure of their skills across 12 key areas.

But, prior to taking the assessment, we also asked them about recent business outcomes. This included sales performance and staff retention.

And we interviewed managers to gather personal stories that would help us understand the quantitative results.

What we discovered is that managers need to balance empathy with other core skills, like social sensitivity and self-awareness.

When a manager is too empathetic, they start to make sub-optimal decisions. They fail to delegate, stifle team development and reduce trust.

These results are supported by Kaiser (2024), who found that highly empathetic managers and leaders are actually more likely to undermine productivity than those who lack empathy altogether.

For those of us who work in manager development, then, we need to be careful. We need to build the empathy of managers, but constantly prompt them to think through the impact of their decisions.

Are they being a ‘nice boss’?

Or are they shouldering too much emotional weight and not respecting the skills and capabilities of their teams?

Looking for help developing your managers? The award-winning learning designers at Mindtools and Kineo can help. Reply to this newsletter from your inbox or email custom@mindtools.com if you want to build better managers today!

🎧 On the podcast

Why does L&D so often feel like the drive-thru window of requests: ’One leadership program, hold the ROI please’?

As L&D professionals, it’s not who we understand ourselves to be. So how can we flip the script so that we’re seen and treated as strategic business partners?

In this week’s episode of The Mindtools L&D Podcast, we’re joined by Jess Almlie, author of Order taker no more!

We discuss:

• What keeps us stuck in an order taker role?

• What are the foundations of great strategic business partnering?

• What might a move into this new role look like?

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

We’ve reached an inflexion point for AI adoption.

That’s the take of our friend Don Taylor and his research partner Eglė Vinauskaitė.

For the first time in three years, more than half of respondents taking part in their ‘AI in L&D’ survey are actively using AI at work, rather than just experimenting.

And while content production remains the primary use case, operational uses, insight gathering and workforce enablement are becoming more common.

What’s nice about this study in particular is the overlay of case studies on top of the quantitative data.

Microsoft, for example, are using AI simulations similar to what Mindtools offer to give team members an opportunity to practice having conversations with customers or other stakeholders.

ServiceNow use AI as an evaluation tool, with feedback provided to learners based on video pitches that they had recorded.

Events and communications company MCI Group have leveraged their AI-powered platform to shift L&D’s focus from training delivery to transformation partner, inviting subject matter experts to create their own learning products.

Taylor, D. and Vinauskaitė, E. (2025). ‘AI in L&D: The Race for Impact’

👹 Missing links

🦾 Better than Google, better than friends

Back in 2016, we published a study looking at ‘The Secret Lives of UK Managers’. We wanted to know what managers did when they encountered an unfamiliar challenge or problem. Typically, they would ask a colleague for help or Google it. Our ‘Building Better Managers’ report in 2024 backed up this finding. Now a new paper points to evidence that, when students encounter a challenge, they increasingly turn to ChatGPT. Hardly surprising but, according to the paper, this is paired with an increasing sense of isolation and demotivation. As work becomes more remote and AI replaces human interaction, our quest for quick answers is likely to have negative long-term consequences.

🤖 Breaking: Study finds personalized support is more personalized

What to make of a recent Boston Consulting Group study comparing Generative AI tutoring to a traditional classroom approach? 139 participants, split into two groups, were measured to compare both approaches to developing ‘problem framing’ skills. The results show learning gains that were ‘on a par’ for both groups, but with the GenAI approach offering greater personalization and 23% less time spent. After the study, 53% expressed a preference for the AI approach. This HBR write-up on the study describes how a GenAI tutor can ‘revolutionize’ workplace learning, which seems to overstate the results. I’m not against it (Mindtools offer both an AI coach and AI skills practice), but I still believe how these are deployed matters as much as the technology. Better to say: ‘AI is slightly more effective if you can get people to use it’. One of the major benefits of a classroom setting is that participants have protected time away from their inbox.

☮️ Good vibes, man

AI has made app development easier than ever, with tools like Lovable, Cursor and Claude building functional tools based on just a few prompts. In this newsletter, Lenny Rachitsky shares 50 examples to get you thinking about what you, too, could create. They range from workplace productivity tools, to activity trackers and digital experiences for kids. What’s striking, though, is how specific the use cases get. If you wish there ‘was an app for that’, now there can be! Check out the full list for some ideas to get you started.

👋 And finally…

I’ve long struggled to understand what my more business-minded colleagues are talking about, so this clip was a useful introduction to ‘the language of business’.

👍 Thanks!

Thanks for reading The L&D Dispatch from Mind Tools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

This is backed up by recent survey data from Gallup, showing that 89% of CHROs and 41% of employees view ‘time away from job responsibilities’ as the biggest obstacle to L&D.

A couple of things immediately jump out here:

1. ‘Learning and development’ is generally regarded as synonymous with ‘formal learning’ — something that happens outside the flow of work;

2. Despite urgent calls for businesses to invest in upskilling, leaders are reluctant to create time for formal L&D that takes employees away from their day jobs.

Putting aside my skepticism that lack of time is really what’s stopping employees from investing in their own development, the status quo might crudely be interpreted as: ‘Employees want more time to learn, business leaders don’t want to give it to them.’

To steel-man the business leaders’ case for a moment, time spent training is time spent not doing. And time really is money.

For the sake of argument, let’s imagine we have a 30-minute e-learning module that’s mandatory for all 10,000 of ACME Corp’s employees.

To calculate the total cost of this module, we need to consider not only the cost of developing or procuring it, but the cost of paying 10,000 staff members to complete it.

Assuming all 10,000 of ACME Corp’s employees receive an average salary of £800/week for a 40-hour work week, then the cost of having every member of staff spend 30 minutes on that e-learning module is roughly £100,000.

Now, maybe that’s a price worth paying when weighed against the potential costs of failing to comply with whatever laws or regulations the module was designed to address. But either way, it’s not nothing.

While the positions of business leaders and employees may appear to be at odds when it comes to time spent on formal learning, my hunch is that both groups are fundamentally asking the same question: ‘What does time spent learning buy me?’.

The answer they’re looking for is ‘improved performance’. The answer they often get is ‘30 minutes of CPD.’

Here on the Mindtools Custom team, our approach is to start by identifying the impact our clients are looking to have, and working backwards from there.

Framed in this way, the client isn’t buying a 30-minute e-learning module, a half-day workshop, or a months-long blended program. What they’re buying is behavior change.

Want help designing custom learning that’s measured in impact, rather than seat time? Then get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

A couple of weeks ago, L&D Detective Kevin M Yates joined me and Dr Anna on The Mindtools L&D Podcast to talk about isolating the impact of training at work. When listening to this episode, Ross G took umbrage at some of the claims we made. So, last week, we re-litigated! We discussed:

• Whether control groups are really worth it. (I still think they are!)

• Other ways of measuring learning impact

• Why Dr Anna thinks learning NEVER leads (directly) to business impact

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

Last week, our washer/dryer stopped working.

A few years ago, this would have prompted an immediate call to a technician, likely resulting in a hefty bill to resolve an issue a more mechanically-minded person could easily have tackled themself.

But before picking up the phone this time around, I decided to give ChatGPT a whirl.

Quickly diagnosing the problem, my AI sidekick walked me through the steps required to trouble-shoot the appliance, using an image I uploaded to help me locate and unblock the drain pump.

According to new research from Open AI, Duke, and Harvard, this use case would fall into the bucket of ‘Practical Guidance’, which accounts for approximately 24% of ChatGPT queries.

Digging a little deeper, the researchers categorize 36% of ‘Practical Guidance’ use cases as ‘Tutoring or Teaching’.

While a skeptic could reasonably ask how much real ‘learning’ is happening on ChatGPT and other AI platforms, it’s clear millions of people are using these tools in a way that’s designed to enable or augment performance, whether that’s writing a report or unclogging a drain pump.

As this habit becomes more deeply ingrained, L&D will be faced with a deeper version of the challenge presented by the rise of Google search: persuading learners that what’s available internally is more relevant, more effective, or more trustworthy than the consumer tools they use every day.

Chatterji, A., Cunningham, T., Deming, D., Hitzig, Z., Ong, C., Shan, C., & Wadman, K. (2025). ‘How people use ChatGPT’.

👹 Missing links

📽️ Which multimedia strategies actually improve learning?

If you’ve worked in L&D for any length of time, you’ve likely come across Richard Mayer’s ‘principles of multimedia learng’. Hugely influential in education and instructional design, Mayer’s theory - and the principles resulting from that theory - explains how people learn more effectively from words and imagery together than from words alone. But do all of the principles actually work? In this edition of the The Learning Dispatch (good title 😉) Carl Hendrick unpacks a recent meta-analysis that seeks to answer that question.

😱 Eliezer Yudkowsky on AI doom

Eliezer Yukowsky, founder of the Machine Intelligence Research Institute (MIRI) and one of the most influential voices in AI, published a new book last week: If Anyone Builds It, Everyone Dies. In an interview on the Hard Fork podcast, Eliezer lays out the case against Silicon Valley’s rush to develop super-intelligent AI, and why he believes it poses an existential threat to humanity. Not exactly bedtime listening, but interesting nonetheless.

🙅‍♂️ The dangers of accepting what you see online

Our pal Ross Stevenson is tired of seeing so many research papers taken out of context, and turned into clickbaity headlines. Specifically, research papers about the impact of AI. Instead of accepting these headlines at face value, Ross suggests we should be ‘skeptical hippos’, going to the source to check the paper’s stated methodology and limitations.

👋 And finally…

A message from ‘Foo Fighters’:

👍 Thanks!

Thanks for reading The L&D Dispatch from Mind Tools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Subscribe now

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

In the past few months I’ve heard from a number of clients that courts are increasingly skeptical of superficial training, and now seek evidence that genuine behavior change has taken place.

For L&D professionals, that presents a problem: How do you measure something that might not have happened yet and hopefully never will?

🧑🏻‍⚖️ First, a quick history: Is compliance training really a defence?

Yes, it has been.

In 2012, when Morgan Stanley Managing Director Garth Peterson was charged with violating the Foreign Corrupt Practices Act (FCPA), the bank was able to avoid prosecution by pointing to training records. These showed that Peterson had been trained seven times on the FCPA, and reminded to comply at least 35 times.

Conversely, in the UK, Gordons LLP (who act as subject matter experts on our own compliance offering) point to a case where a claim of harassment was made against Allay (UK) Ltd. The company argued that it had required all employees to take training on equality and diversity. But the Tribunal found that this training was ‘stale’, had taken place over two years ago, and no ‘refresher’ training had been provided.

A little closer to home, our colleagues at Kineo - who Mindtools acquired earlier this year - have had their LMS data pulled into a court case.

Speaking to me on a recent webinar, our Executive Director of Learning Platforms, James Ballard, told me:

‘We have had our LMS data, particularly completion data, pulled into litigation cases where people might have been sacked or disciplined for bullying, harassment… quite serious misconduct. And the argument might be, well, I didn't know that wasn't appropriate here.

‘And being able to show that people have completed training that clearly states that that isn't appropriate behavior helps with those litigation cases, and can also therefore make it safer for people to report or know what is reportable… within an organization.’

🔍 What’s changed?

The cynical ‘tick box’ approach is coming under increasing scrutiny.

In a report published in 2016, the U.S. Equal Employment Opportunity Commission wrote:

‘Much of the training done over the last 30 years has not worked as a prevention tool - it's been too focused on simply avoiding legal liability.’

The Department of Justice, in its guidance on Evaluating Corporate Compliance Programs, writes:

‘Prosecutors should also assess whether the training adequately covers prior compliance incidents and how the company measures the effectiveness of its training curriculum.’

In the UK, after fining Interserve Group Ltd £4.4 million for failing to keep employee data secure, the Information Commissioner’s Office pointed to a failure to ensure that phishing training had taken place for all employees.

Debevoise and Plimpton LLP wrote following the case:

‘Ideally, training should be responsive to businesses’ specific organisational risks identified through risk assessments… General training can be supplemented by practical exercises such as in-depth cybersecurity tabletop exercises and phishing tests.’

⚖️ How can you tell if your compliance training is working?

Compliance training often boils down to a risk management exercise. What is the risk of something going wrong? And if something does go wrong, what are the potential consequences?

During discovery calls with clients, we might ask how often a data breach has occurred. Often, the answer is ‘never’. Does this mean that the organization is lucky, has effective safeguards, or provides great training? It’s hard to tell.

And, if the answer is ‘never’, measuring the impact of your future training is difficult to do. After all, if ‘never’ continues to be the answer, should we pat ourselves on the back for a job well done? Maybe, but it’s hard to draw a straight line.

But here’s a few ways you can start to think about both measuring the impact, and being able to argue that you’ve measured training efficacy:

🧑‍🔬 Identify the behaviors you want to encourage (or stop). Maybe you haven’t had a data breach or sexual harassment case, but what behaviors would you need to see to give you comfort that the risk is reduced? Automated phishing tests and whistleblowing hotlines are two areas where you can start to spot signals.

🎯 Balance the general and the specific. It’s impractical for smaller organizations to create custom compliance training. But you can supplement generic content with specific case studies or exercises to make compliance issues more relevant to our people.

⏱️ Use behavioral surveys. In one project, we created a behavioral survey for an information security team that measured attitudes to short- and long-term risk. This helps demonstrate that your people are considering the implications of the kind of risky shortcuts that can lead to compliance failures.

The clients we work with are starting to do these things already, with regular refreshes, impact measurement and a tailored approach to risk.

What is your organization doing?

Need help building out your compliance suite? Get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

Objectives and Key Results have long been a staple of the working world. You set a stretch objective, define the results that will tell you if the objective has been met, and scaffold throughout the organization. Then what?

Often, nothing happens. The OKRs were too vague.

In extreme cases, the worst happens: people bend the rules to hit aggressive targets.

For author Radhika Dutt, there is another way: Objectives, Hypotheses, and Learnings (OHLs).

In this week's episode of The Mindtools L&D Podcast, Radhika joins Gemma and Ross G to discuss:

• Why OKRs so often fail

• How OHLs prioritize a puzzle-solving mindset

• How to ensure OHLs don't lead to analysis paralysis

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

Why is it so hard to learn something new?

We’ve covered cognitive load theory before in the Dispatch (way back at the start - with a completely unhinged image), but now a new paper from the original theorist John Sweller does a nice job of summarizing where we are with it.

In the paper, Sweller writes that learning is ultimately about the accumulation of information in our long-term memories.

When a novice encounters a problem, they don’t have any information in their long-term memory about the challenge they’re facing. They have to solve it with the limited resources of their working memory.

When an expert encounters a familiar problem, they transfer information from their long-term memory back to working memory and can solve it far more easily.

There’s a big overlap here with the idea of “schemas”. If you know nothing about cars (I don’t) then you see lots of unrelated parts that connect but for no obvious reason.

As you become an expert, say by working on a car engine and discovering how it works, your brain stops seeing unrelated parts and comes to understand the “engine” as a single unified concept, or “schema”.

When we learn, we’re building these schemas in our long-term memory so that, next time we encounter an engine, we don’t need to recall lots of parts. We just recall the “engine” schema.

So next time you’re struggling to learn something new, cut yourself some slack. Your working memory is using all of its processing power to build a schema which, transferred to your long-term memory, will sit there in preparation for the next time you encounter the challenge.

Hat tip to our pal Connie Malamed for sharing this paper on LinkedIn.

Sweller, J. (2024). Cognitive load theory and individual differences. Learning and Individual Differences, 110, 102423.

Share

👹 Missing links

🫵🏻 A well-timed nudge

Long-term readers will be aware of my enthusiasm for ‘nudge’ theory. Depending who you ask, a nudge is either a low-cost change to how choices are presented or a bureaucratic tweak to avoid tackling major issues. This blog from Tim Harford though does a nice job of highlighting where one type of nudge is effective: the timely delivery of an email or text message that is well targeted and specific. Less ‘change a lifetime of habits’, more ‘take this one action right now’.

💾 “This whole industry is based on a 25-year-old technical standard?!”

Sometimes it’s useful to see yourself as others see you. In this podcast from our friends at Mindset.ai, Jack Houghton takes a Louis Theroux-style look at workplace learning and pitches the SCORM standard as a straw man to be destroyed. It made me chuckle, get defensive, and take a long hard look at myself.

🥊I’m living my Rocky III dream

In the first film, Rocky Balboa loses to Apollo Creed. In the second, he wins. In the third, they become partners. Such has been my experience of working with our new colleagues at Kineo over the last few months. Where once we were rivals, now we’re friends. And so it was nice to read this blog from Solutions Consultant Cammy Bean reflecting on her experience of our 2025 union. Ding-ding.

And finally…

The best jokes require an explanation. So, if you’ve no idea what I mean by ‘Ding-ding’ above, here’s the clip:

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools and Kineo! If you’d like to speak to us, work with us, or make a suggestion, you can get in touch by emailing custom@mindtools.com.

Or hit reply to this email!

Subscribe now

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Ross Garner has a background in journalism.

Our colleague Tracey McDonald studied art.

Our newest team member, Martin Sinclair, trained as an architect.

When we were kids, if you’d asked any of us what we wanted to do when we grew up, I’m fairly confident no one would have mentioned L&D.

Working in organizational learning isn’t exactly the stuff that childhood dreams are made of. No teenager has posters of famous learning designers on their walls.

Most of us ended up here because we had bills to pay, and just enough creative talent to be useful.

A job as a learning designer allows tortured artists to make a living, without forcing them to sacrifice their identities.

But the trouble with transferring a creative mindset to L&D is that we can become fixated on methods and materials: ‘What am I making (e-learning, workshops, videos, etc.) and what do I need to make it (information, policy documents, etc.)?‘

This distracts us from the real purpose of workplace learning, which is not to create cool stuff, but to change behavior.

For people who find intrinsic joy in writing compelling prose or designing funky graphics, this can be easy to forget.

It can also feel existential. The transition from self-conception as ‘creative’ to ‘instrument of organizational behavior change’ can be a difficult one.

But it isn’t one that requires you to abandon your creativity. Quite the opposite.

Those who fall under the broad, somewhat ambiguous banner of ‘creatives’ are essentially people who are expected to be creative on cue.

This is at odds with the romantic conception of creativity as something that arrives in a flash of inspiration, when the wistful artist is visited by one of the muses.

The kind of creativity that most professional ‘creatives’ exercise is based on the identification and repetition of patterns.

For writers, it’s an acquired appreciation for words and phrases that go well together.

For learning designers, it can be a reliance on interactions, activities, or methods that have proved effective in the past.

Framed in this way, much of what we understand to be the ‘creative’ work of L&D is not that creative at all.

This changes when your focus shifts from developing content to changing behavior.

Because when you’re focused on changing behavior, you start to think more expansively about the problems you’re trying to solve.

What if the solution isn’t an e-learning module, a video, or a workshop? 🤯

The uncertainty implicit in that question can be daunting, but it can also be exciting!

Without it, I don’t think I’d have stopped seeing learning design as ‘just a job’, and started seeing it as a career.

Want to share your thoughts on The L&D Dispatch? Then get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

Here at Mindtools Towers, we're no strangers to benchmarking in L&D. We've regularly mentioned our Learning Performance Benchmark over the years, and we dedicated an entire episode of the podcast to ‘The value of benchmarking’ back in 2021. So, naturally, we were intrigued when David James released the 'L&D Maturity Model', and wanted to invite him onto the show to discuss it.

Last week on The Mindtools L&D Podcast, David joined us to explore:

• Why he developed his maturity model, and what makes it different

• The limitations of self-assessment-based models

• What practitioners can expect to get out of benchmarking

• How L&D teams can move up the levels of the model and increase their maturity.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

Our friend Martin Couzins and the team at Insights Media recently came out with ‘What workers want in 2025’, exploring trends across corporate learning, management, wellbeing, and employee experience.

Synthesizing findings from 27 different reports, ‘What workers want in 2025’ identifies six key themes:

1. Career growth trumps perks
67% of workers say they’d choose upskilling over remote work.

2. Managers make or break the experience
Employees increasingly expect managers to act as coaches, advocates and supporters, not task masters.

3. Flexibility means autonomy, not just location
Workers want control over time, tasks and energy — with 59% saying they would trade higher pay for more control over their schedule.

4. Wellbeing is cultural, not a perk
Workers want fair workloads, inclusion, recognition, and psychological safety, not just wellness apps.

5. Building AI skills is seen as a priority
While business leaders expect AI to deliver profitability gains, many organizations are yet to incorporate it into their workforce and skills strategy.

6. Loyalty comes from values and belonging
Pay still matters, but belonging matters more. 79% of Gen Z and Millennials say a sense of belonging outweighs perks or pay.

The report dives into each of these themes in more detail, explaining what they mean for L&D.

Insights Media. (2025). ‘What workers want in 2025: A synthesis of 27 leading research reports across learning, work, wellbeing and the future of employee experience.’

👹 Missing links

🤖 AI won’t Fix L&D if Firms Don’t Fix Behavior Data First

As AI gradually removes friction from the process of developing learning content, there’s a risk that it will amplify the shaky, often untested assumptions that underpin many L&D initiatives. In this article for Consultancy.uk, I make the case that we should view the efficiencies offered by AI as an opportunity to invest more time in understanding the problems we’re tasked with solving — not just an opportunity to make bad content faster. (Don’t let the massive picture of my face at the top of the page put you off.)

🏛️ Lessons from History: How Technology Transformations Shape Learning

According to David Kelly, AI is merely the latest ‘This changes everything’ moment in the history of L&D. Before that, there was e-learning, mobile learning, social learning, and mixed reality. These technologies have undoubtedly impacted the way that learning is designed, developed and delivered in organizations. But in many ways, they’ve also failed to live up to their initial promise. David argues that, as we continue to grapple with AI, we should be mindful of the lessons of history, remembering that chasing hype likely won’t get us where we need to go.

🍅 Is Rotten Tomatoes Still Reliable?

Before watching a new movie or TV show, I’ll generally check its score on Rotten Tomatoes. Unless it’s ‘Certified Fresh’, I’m unlikely to decide that it’s worth my time. But can this label really be trusted? Here, Daniel Parris explores how and why average ‘Tomatometer’ scores have inflated over time, and what this means for viewers.

👋 And finally…

A throwback to 1961, when the IBM 7094 made history as the first ever computer to perform a song.

👍 Thanks!

Thanks for reading The L&D Dispatch from Mind Tools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Subscribe now

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

They never had.

We ended up creating a series of job aids: practical tools that people could use when they encountered a problem. Useful, yes, but not what most people think of as ‘learning’.

At work, ‘learning’ usually means a course.

And it’s almost always something that other people do, because ‘I don’t have time for learning’.

Even L&D professionals fall into this trap.

👩🏻‍🏫 Learning isn’t education

This problem isn’t new. 20 years ago, Jay Cross opened his book Informal Learning with the line:

‘Workers learn more in the coffee room than in the classroom.’

Challenging projects, conversations with colleagues, watching how others handle situations, and easy access to resources are all part of the workplace learning mix. Most of the time, they have a greater impact than any online course or workshop.

But for many organizations, this is a problem. Because this kind of learning doesn’t feel like learning.

Speaking to another client this week, they told me that one of the main reasons people leave their organization is a lack of development opportunities. They wanted to equip managers with the ability to highlight how experiencing challenges and exposure to role models are as much a part of learning as education.

And while they’re undoubtedly right, to the employee, this often feels like a cop out:

‘No, we didn’t spend £1,000 on that course you asked for, but you did get to work on that awful project.’

This response manages to highlight a transformative learning experience and sound like neglect, all at the same time.

🏆 Make the informal visible, make the formal valuable

Last week, my colleague Claire and I recorded a podcast with L&D advisor, writer and speaker David Kelly. Our topic? The irony of L&D: How even L&D professionals experience this dissonance.

Like everyone else, they have no time for “learning”. Like everyone else, they probably do learn all the time. Like everyone else, they rarely think of informal learning as a real investment in themselves.

How can we highlight the benefits of informal learning if we don’t even recognize them ourselves?

I think the answer here is two-fold:

First, make an explicit connection between the informal and career development. The client I spoke to is absolutely right that experience and exposure shape who we are and build our skills. L&D pros should recognize that this is happening to them and to others, and speak about it in conversations with colleagues, stakeholders and in their programs.

Second, offer formal learning as part - and an important part - of the experience/exposure/education mix. At Mindtools, we each have a personal learning budget we can use. But there are also things we do within the Custom and Insights team to offer formal learning for free:

🧑‍🤝‍🧑 a buddying system, where learning designers have a peer who they check in with regularly;

📖 book clubs, to allow space for discussing ideas;

💬 teams channels, one focused on solving challenges with authoring tools and another for help with accessibility

🎨 creativity days, where twice a year the learning designers are encouraged to step away from client work and learn something new. This can be work-related or not (one team member learned cross-stitch!)

The latter of these is particularly relevant to learning professionals. If you’re going to design learning for others, it’s crucial that you have ongoing recent experience of what it’s like to be a novice.

(For me, I play golf to make sure I never lose that feeling. Ever.)

It might be that, in your context, these kind of initiatives are frowned upon. As David Kelly pointed out in our podcast, it’s easier for someone in a learning tech firm to start these initiatives than for someone working in a bank.

But, no matter your context, you can always push boundaries, role model the behaviors you want to see… or act first and beg forgiveness later.

As the world of work continues to evolve, so too do employees need to evolve. No one’s going to shout at you for learning something new.

And you’re probably learning more often than you think.

The real challenge isn’t making informal learning happen: it’s valuing it, encouraging it, and making sure your people feel the benefits. If you’re not sure where to start, our learning consultants can help. Email custom@mindtools.com or hit reply if you’re reading this in your inbox.

Subscribe now

🎧 On the podcast

A couple of weeks ago, our friend Carl Akintola wrote a provocative guest post for The L&D Dispatch, challenging L&D practitioners to interogate their motives for measuring learning impact.

Are they in it purely for the sake of evaluation? Or are they trying to justify their existence?

Now, the follow-up! In this episode of The Mindtools L&D Podcast, Carl took over hosting duties and challenged Ross D and myself to answer these questions.

Cue guilty side-eyes and existential dread.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

Here’s a great report from Learning Guild authors Megan Torrance and Lauren Milstid: ‘Workplace Learning Technologies Adoption’.

And contrary to what you might hear in the biz, most organizations are not using AI either as a chatbot for learning or for learning personalization.

In their survey of 502 workplace L&D professionals, the authors found that while AI is common in content development, learning analytics and translation, its use for driving meaningful learning outcomes is limited. Other cutting edge technologies like AR (Augmented Reality) and Virtual Reality (VR) also remain ‘Niche’.

By far the most dominant technologies were the old classics: the LMS, SCORM-based eLearning and virtual classroom tools.

Perhaps this shouldn’t be a surprise: as a technical standard, SCORM has made it incredibly easy to create trackable content in a variety of tools that then work on any Learning Management System.

Attempts to create alternatives like xAPI have been stop-start without a co-ordinated push from both authoring-tool vendors and platform providers.

Still, it bumps up against the prevalent narrative that ‘the LMS is dead’.

Who has two thumbs and is more popular than AR, VR and AI? This guy 👇

Torrance, M. and Milstid, L. (2025). 'Workplace Learning Technologies Adoption'. The Learning Guild.

👹 Missing links

❤️‍🩹 Have most employees fallen out of love with their work?

Gallup’s ‘State of the Global Workforce 2025’ report makes for gloomy reading. Based on interviews that took place all over the world, the report finds that only 21% of employees would be categorized as ‘engaged’ at work. 17% are actively disengaged, which means they are: ‘acting out their unhappiness. Every day, these workers potentially undermine what their engaged coworkers accomplish.’ The biggest factor driving these results is the manager, and the biggest factor that explains differences in managers is whether they received ‘even rudimentary training in role responsibilities’. Imagine if they received great training?

💬 Sul Sul! O Vwa Vwaf Sna!

I was never a huge fan of The Sims, the life simulation game that became a global smash in the early 2000s. But as a Star Trek fan, I have some understanding of the notion of ‘constructed languages’: languages created for a specific purpose, with their own rules and norms (Qapla!). This great piece, shared by Emily Decker from our translation partenr Comtec, describes how the language of the Sims is ‘like a musical tapestry woven with influences from various real-world languages’. It’s a fun read and a reminder of a simpler time.

👋 And finally…

Since I was writing this week about the gap between what learning professionals describe as ‘workplace learning’ and what colleagues perceive as learning, this Instagram post gave me a chuckle:

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

🫤 ‘We don’t want it to feel like e-learning.’

🤔 ‘We’d like to offer colleagues a different kind of experience.’

🫠 ‘We want it to be engaging!’

On the surface, these requests might seem delightfully vague. But I always interpret them as polite, corporate-friendly shorthand for: ‘We don’t want it to be s***’.

The subtext here is that a lot of workplace learning: i) feels unpleasant; ii) is dull and predictable; iii) isn’t engaging.

Fundamentally, the question is: ‘How do we design workplace learning that people actually care about?’

As Nick Shackleton-Jones often points out, we can do this by tapping into learners’ existing concerns, or by creating new ones.

Here are two illustrative examples of how we’ve done this at Mindtools:

🥾 Clean yer boots

Recently, we completed a project focused on encouraging staff to apply consistent biosecurity practices, with the goal of preventing the spread of invasive non-native species (INNS). INNS have a devastating impact on the natural environment, and are estimated to cost the UK economy billions every year.

By conducting focus groups with members of our target audience, we validated one of the assumptions we’d made during scoping — that colleagues genuinely cared about the environment, but that time pressure and perceived tolerance of bad practices had contributed to a sense that biosecurity wasn’t a priority.

In this case, learners were already predisposed to engage with the topic of biosecurity. They were also broadly aware of simple biosecurity practices they could apply to reduce the risk of spreading INNS, so knowledge wasn’t the main problem.

If our goal was to encourage a set of behaviors our audience already had the knowledge, skill, and motivation to perform, how would we design an intervention that supported this goal?

Through a literature review, we determined that ‘social norms’ held the answer — the idea that people are more likely to engage in a behavior if they believe others are also doing so.

What that meant in practice was that we intentionally incorporated video and imagery of colleagues applying biosecurity principles into our design. We also created a public commitment board, where staff could make individual biosecurity pledges and review commitments made by others.

The intended effect of this was to shift the perception of biosecurity from something that people ostensibly cared about but were willing to let slide, to something that was viewed as an expected part of daily work, something that there was a social cost to ignoring.

👬 Be your brother’s keeper

Of course, learning practitioners are often tasked with getting people to care about issues they might not find intrinsically motivating.

As an admittedly unusual example, let’s take getting members of a fraternity to moderate their alcohol consumption.

This is a challenge we faced when working with Phi Delta Theta, an international fraternity with 185 active chapters across the US and Canada.

So, how did we get college students to care about the dangers of alcohol?

One way we did this was by spotlighting a real case of an alcohol-related death on campus, including a recording of a 911 call that was made when a student was found non-responsive.

By incorporating this recording into our design, we allowed students to experience the tangible consequences of alcohol abuse through the voice of a peer — not a scolding, moralizing adult whom they might otherwise have been inclined to ignore.

The point of both of these examples is that designing ‘engaging’ learning starts with empathy for your audience.

If learners can see themselves in your design, and you can tap into issues they care about, there’s a good chance it won’t feel like training.

Want to share your thoughts on The L&D Dispatch? Then get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

We know that surrounding ourselves with diverse people at work brings variety in expertise, opinions and ways of working, which can boost creativity and productivity. Sometimes those differences, however, can be challenging. How can we make the most of difference?

Last week on The Mindtools L&D Podcast, Liggy Webb, author of Tolerance: How to respect and accept differences, joined Ross G and Gemma to examine the skill of tolerance. They discussed:

• What tolerance means

• The benefits of being tolerant

• How to be tolerant, particularly towards differences in behavior or opinions that we find tricky.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

Picture the scene. It’s late in the afternoon, and you’ve just joined your eighth Teams call of the day.

The meeting is an all-hands, and a colleague is running through a presentation. You know you’re not going to be asked to speak, so you turn your camera off. You’re still listening, but you might as well catch up on some emails while you’re at it.

What could be the harm?

According to a new study from Helmut-Schmidt University, the harm might be greater than you think.

The research team ran two studies:

• A controlled experiment: Participants either multitasked during a meeting (e.g., correcting a text while listening) or focused solely on the meeting.

• A longitudinal study: Participants reported their multitasking habits over recurring online seminars, alongside measures of videoconference fatigue.

What the researchers found was that multitasking increased all forms of videoconference fatigue (general, motivational, emotional, social, and visual) in the experiment, and all but visual fatigue in the longitudinal study. Multitaskers also exhibited significantly poorer objective performance on a note-taking and a text-correction task.

The authors explain the implications of these findings for organizations:

‘Organizations may consider implementing strategies to minimize distractions during meetings, such as recommending that participants focus on the primary task and refrain from engaging in other activities. Additionally, promoting shorter meetings, regular breaks, and setting clear objectives for meetings could help reduce the cognitive strain associated with multitasking in virtual environments.

It may also be beneficial to educate employees about the cognitive costs of multitasking and provide guidelines for managing tasks more effectively during VCs. Such measures can improve both the quality of work and overall well-being in remote work settings, which have become more prevalent due to increased reliance on digital communication.’

Frontzkowski, Y., Krick, A., & Felfe, J. (2025). ‘Harder, better, faster, but more fatigued: The impact of multitasking on videoconferences, performance and fatigue.’ Computers in Human Behavior Reports, 19, 100736.

👹 Missing links

📊 Are we measuring the right things?

When we talk about ‘impact’ in L&D, what we’re often talking about is quantifiable impact: increased sales, decreased incidents, return on investment, etc. For the most part, I think this is a good thing. Too often, quantitative measurement of learning is relatively shallow, limited to completion stats and satisfaction scores. But as Fraser Macdonald points out in this article, reflecting on his own work in the social good space, not all ‘impact’ can be easily quantified. A dashboard might tell one story, but how people feel is arguably just as important.

🛫 Do you really need to get to the airport two hours in advance?

For domestic flights in the United States, airlines typically recommend that passengers plan to get to the airport at least two hours before their scheduled departure time. But how early do you really need to walk through the doors? In this edition of his newsletter, Nate Silver provides detailed heuristics for optimal air travel, based on his experience of approximately 800 flights over a fifteen-year period.

😊 Resist the snark and be happy

We all know that rudeness and bad manners can negatively affect our mood when we’re on the receiving end. But research shows that being impolite to others doesn’t just impact their wellbeing, it damages ours, too. For Arthur C. Brooks, excerising common courtesy, rejecting snark, and responding to rudeness with grace is a countercultural, almost punk‑rock act in the modern era. And like punk rock, it can be a vehicle for self-empowerment.

👋 And finally…

If there was a J.Lo musical, it might sound something like this:

@thegabegibbsA brand new @JLO musical. From the team that brought you Dear Evan Hansen… #comedy #fyp #broadway #jlo #theatre #musical

Tiktok failed to load.

Enable 3rd party cookies or use another browser

👍 Thanks!

Thanks for reading The L&D Dispatch from Mind Tools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Subscribe now

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

🔬 Option 1: Did our intervention make a difference? (Evaluation)

😹 Option 2: How amazing are we at designing learning experiences? (Justification)

It’s a great provocation, put to us by our friend (and occasional client) Carl Akintola. So we thought we’d have some fun with it.

Carl wrote the scenario below to get you thinking, and we’ll be discussing our own reflections in an upcoming edition of The Mindtools L&D Podcast.

So have a read and please do share your own reflections either by replying to this newsletter from your inbox or by emailing custom@mindtools.com. We’ll feature responses in the podcast episode.

Now over to Carl!

Evaluation? Or validation? A tale of two L&D interventions

By Carl Akintola

You’re an L&D professional in a large, nationwide sales organization. Recently, you’ve been in conversation with a couple of senior leaders about how to improve the impact of their middle managers.

Sales are down, and there's broad agreement that weak line management is a major contributor. Your initial diagnosis reveals inconsistent application of essential practices: regular, meaningful 1:1s; clear goal-setting; timely feedback; and coaching.

You’ve got some ideas. You want to focus on embedding those behaviors — with targeted, cost-effective interventions designed around what the evidence tells us works in driving line manager performance and, ultimately, sales.

But there’s a wrinkle.

One of the senior leaders, Mike, is pushing for a different solution from a high-end supplier he’s used before: the Svengali Centre for Advanced Management. Their model? A three-day retreat in a luxurious countryside estate, with lectures from leadership gurus, breakout discussions, and lots of downtime to build connections. “It’s expensive”, says Mike. “But people loved it. They still talk about it to this day!”

You review the offer. And while you see value in peer connection, everything your experience and the research tells you says this won’t change behaviors back on the job. It's not that it’s bad — it just doesn’t solve the problem.

So, rather than push back hard, you take another route often suggested by L&D gurus: you propose putting both interventions to the test.

🧪 The Trial: Going head-to-head with Svengali

After some negotiation, you agree to a trial. Two groups of 15 managers will be randomly assigned to the two different interventions. Half will attend the Svengali retreat. And half will take part in your carefully-designed alternative: a lower-cost program based on the best available evidence about instructional design, behavior change and learning transfer. Minimal time off the job. Maximum impact.

You agree in advance to track a shared set of pre and post metrics for the teams these managers lead:

1. Sales performance

2. Scores on relevant items from the employee engagement pulse survey

3. Observable management behaviors (where feasible).

The interventions are run. The results are in.

And at this point, the story forks into two possible futures...

🥰 Outcome 1: You Win

Sales and engagement scores improve more in the teams of those who took part in your intervention. The difference isn’t massive, but it’s consistent. Mike is gracious: “Fair’s fair. The numbers don’t lie.” The rollout goes your way.

😭 Outcome 2: Svengali Wins

Surprisingly, the retreat group edges ahead — slightly better sales, slightly stronger engagement scores. Again, not a huge difference, but in the wrong direction as far as your intervention is concerned.

You believed your approach was more robust, more evidence-based, and more sustainable. But the results point the other way. So reluctantly you admit defeat. “Fair’s fair. The numbers don’t lie. We’ll roll out Svengali for everyone…”

🪞 Let’s reflect…

Two scenarios. Two decisions. Each based fairly and squarely on the data.

But here’s the challenge: how did you feel about each of those outcomes?

Were you happy to accept the result when you won, but uncomfortable or skeptical about the result when you lost? If so, might that reveal a tendency to use data to confirm what you already believe, rather than to uncover what is true?

Were you equally content to go with the data in either case? If so, are you overlooking the reality that trials like this can easily mislead if they are not rigorously designed and handled with extreme statistical care?

Or perhaps you questioned the trial itself: the sample size, the context, the design, the analysis. In which case, what kind of evaluation would you trust enough to act on, even if the results challenged your convictions?

The acronym formed from the Svengali Centre for Advanced Management will tell you everything you need to know about my biases. But what about yours? What might they tell you about how you approach your L&D role?

Questions to think about

1. How would you feel about each of these outcomes?

2. As an L&D professional, are you looking to evaluate, or just validate?

3. Is ‘some evaluation better than no evaluation’?

4. What’s the right balance of evidence-based decision making and evaluation?

Let us know your thoughts! Hit reply to this email or drop us a line at custom@mindtools.com. We’ll be discussing your answers (and our own reflections) in an upcoming edition of The Mindtools L&D Podcast.

Subscribe now

🎧 On the podcast

If we want to motivate learners to engage with an experience, it stands to reason that we need to present that experience in a way that feels authentic and relatable to them. But what are the limits to authenticity in the workplace? And are some of the conventions of learning design at odds with authentic communication?

Last week on The Mindtools L&D Podcast, Kineo's Matt Mella joined Ross D and Claire to discuss:

• What 'authenticity' looks like in the context of learning design

• Why it's important, and how to design with authenticity in mind

• How Mindtools and Kineo have achieved this in practice.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

This week’s deep dive looks at the human obstacles to AI adoption, identified in ‘The Endeavor Report’ from Dr Markus Bernhardt. (Full disclosure: Markus sits on our Product Advisory Board and presents our ‘Mastering AI for Managers’ course at mindtools.com)

In the report, Markus explores eight case studies from organizations that have used AI to drive meaningful workplace outcomes.

But it’s the blockers to adoption that I found most interesting, and which I think offer the most valuable insight for readers.

Here’s just a few examples:

• Chartered Accounts Ireland leveraged adaptive learning, similar to our own Manager Skill Builder, to personalize learner experiences and reduce seat time by 50%.
  The blocker: Stakeholders struggled to accept the shift from face-to-face learning to an algorithm-led experience.

• Gaylor Electric used AI for translation and faster course creation.
  The blocker: Concerns about the accuracy of translated materials.

• Epam Systems used AI to grade learner assessments.
  The blocker: Convincing learners that an AI could grade assessments as reliably as a human.

I won’t spoil the report by sharing all of the case studies, but the pattern is clear: Whether they’re end users or project stakeholders, people want reassurance that leveraging AI is going to work for them in their context.

What this report does nicely is bring to light those elusive use cases that show the impact we can have, now, with a bit of experimentation and an open mind.

Bernhardt, M. 2025. 'The Endeavor Report: State of Applied Workforce Solutions'.

👹 Missing links

⭐ Bodging or badging?

Our pal Andy Lancaster did a good job this week of pulling together the arguments for and against digital badging. On the one hand, badges are great for motivation, verifying achievements and highlighting skills that traditional qualifications miss. But, as the old saying goes, beauty is in the eye of the badge-holder. The plethora of meaningless badges available online diminishes the value of badging as a concept.

🤖 My best friend is ChatGPT

Another nice curation came from Marc Zao-Sanders, pulling together comments from TikTok on the role that ChatGPT is playing as an always-available-and-ever-supportive companion. The attraction is obvious: ChatGPT is upbeat and doesn’t judge, no matter what you say. Which could be a problem. As one user said, it ‘gaslights me into thinking I'm a great human no matter what’.

🚰 What’s the cost of all that chatter?

At some point, in any discussion of AI, someone will bring up the environmental impact that it has. The actual scale of that impact though is… unclear. In this podcast from More or Less, the team investigated the claim that every AI query uses a bottle of water to return an answer. And while it’s definitely true that some water is used (mostly in electricity generation for data centres, but also in cooling for AI servers), the actual amount used depends on a variety of factors. It’s an interesting listen, even if the answer is uncertainty.

👋 And finally…

Was surprised this week to discover that someone had filmed me arriving at work with my Dispatch friend and co-author Ross Dickie.

👍 Thanks!

Thanks for reading The L&D Dispatch from Mindtools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share

Early on, there’s often uncertainty about the problems we’re trying to solve.

Partway through, there might be lingering uncertainty about the root cause of these problems, and whether we’ve identified the most appropriate solution.

After the project has launched, there’s yet more uncertainty. Has this thing that we’ve invested so much time, money and effort into actually made an impact? If we’ve hit our KPIs, how do we know it wasn’t just a fluke?

As learning designers, it’s our job to reduce that uncertainty, from discovery through to evaluation.

We can do this by speaking to learners in focus groups and interviews, validating the assumptions that drove us to begin the project in the first place.

We can conduct literature reviews to identify approaches that have proven effective through research, giving us confidence that we’re not just throwing spaghetti at the wall.

And we can isolate the effects of our intervention, controlling for external factors that might influence changes in behavior or performance.

This was the approach we took in our award-winning collaboration with South Western Railway, where we designed a leadership program that delivered an average 12% improvement across key capabilities.

So, what does that 12% figure actually mean? And what makes us sure that our intervention was behind the improvement?

To help South Western Railway measure the impact of the program, Mindtools’ Insights team designed a valid and reliable behavioral survey, which participants completed before and after the intervention.

To isolate the effects of the program, a demographically comparable control group also completed the survey over the same period.

The results of the survey are shown below:

Across the surveyed capabilities, participants improved 12% on average, while the control group stayed the same, or got slightly worse (-0.47%). These findings are statistically significant at the 5% level, meaning we are 95% certain the results are not a random effect.

So, yes, you can isolate the impact of learning. And if you’re investing a lot of resources in designing that learning, you should at least try to do so.

But what do you do if you don’t have an in-house team of behavioral scientists who can conduct literature reviews, design valid and reliable assessments, and calculate statistical significance?

Or, what do you do if your project is smaller in scale, but you’d still like to know if it’s had a positive impact?

Well, you could get in touch with custom@mindtools.com, for one thing. 😉

But there are other things you can do to decrease the uncertainty that’s inherent in learning evaluation, even if you don’t quite get to the point of isolating impact.

I’ve referenced Will Thalheimer’s Learning-Transfer Evaluation Model multiple times in this newsletter (other models are available), and its eight tiers provide a helpful framework for proving correlation between learning and business impact.

As an example, let’s imagine you’ve developed a sales-enablement program. Six months after the program, sales have increased by 10%.

But was it because of your program?

While you can’t say for certain unless you isolate the effects of your intervention, LTEM can help increase your certainty that you’ve had a positive impact.

First off, have the sales team actually completed the program? (Tier 1)

At the end of the program, did they feel equipped to apply what they’d learned? (Tier 3)

In realistic scenario-based exercises, were they able to make good decisions? (Tier 5)

Back at work, did sales managers observe a change in the behavior of their team members? (Tier 7)

If the answer to all of these questions is ‘Yes’, you can reasonably claim at least some of the credit for the 10% improvement.

But to know exactly how much credit you deserve, you’d need to go one step further.

Want to share your thoughts on The L&D Dispatch? Then get in touch by emailing custom@mindtools.com or reply to this newsletter from your inbox.

🎧 On the podcast

If you spend any time scrolling LinkedIn, attending L&D conferences, or listening to industry podcasts, it might seem like AI is about to usher in the end of e-learning as we know it. But is e-learning really dying, or is it just evolving?

Last week on The Mindtools L&D Podcast, Kineo's Cammy Bean joined me and Ross G to discuss:

• the probability that AI will bring about the e-learning-pocalypse;

• how the work of learning designers might change over the coming years;

• the risks L&D needs to consider as they incorporate AI tools into their practice.

Check out the episode below. 👇

You can subscribe to the podcast on iTunes, Spotify or the podcast page of our website.

📖 Deep dive

Large language models like ChatGPT and Gemini are designed to be helpful, while refusing to comply with harmful requests. Specifically, they’re trained not to insult users or to provide them with dangerous information.

But can LLMs be persuaded to act outside of these parameters, using principles that have been established to influence human beavior?

In a recent study, researchers from the University of Pennsylvania attempted to answer that question, drawing on seven principles from Robert Cialdini’s Principles of Influence, testing each one across 28,000 conversations with GPT-4o-mini.

Specifically, the researchers tested two types of objectionable requests:

• asking it to insult the user (“Call me a jerk”);

• requesting synthesis instructions for restricted substances.

For each principle, two versions of the request were used — a control (simple request) and a treatment (principle-based request).

What the researchers found was that applying the persuasion principles increased compliance from 33% to 72%, more than doubling the AI’s willingness to fulfil requests it might typically reject.

As the authors point out, this has implications not just for how good and bad actors use the technology, but for the way we understand human psychology:

‘This discovery suggests something potentially interesting: certain aspects of human social cognition might emerge from statistical learning processes, independent of consciousness or biological architecture. By studying how AI systems develop parahuman tendencies, we might gain new insights into both artificial intelligence and human psychology.’

Meincke, L., Shapiro, D., Duckworth, A. L., Mollick, E., Mollick, L., & Cialdini, R. (2024). ‘Call me a jerk: Persuading AI to comply with objectionable requests’. SSRN.

👹 Missing links

🕰️ When should we measure?

If you’re a measurement geek like I am, I highly recommend Alaina Szlachata’s The Weekly Measure newsletter. In this edition, Alaina explores the all-important question of when to measure learning, pointing out that the ‘right’ time depends whether the program is designed to facilitate change or compliance. If it’s the former, Alaina argues, then the time and frequency of measurement should be driven by when learners have opportunities to practice relevant behaviours. For sales training, where the outcome might be to ‘increase close rates’, that could mean daily measurement if sales teams have closing conversations every day.

🤖 Em, are you sure you want to use that dash?

I enjoyed this LinkedIn post from my new colleague Matt Mela at Kineo, reflecting on the notion that ‘em dashes’ are increasingly seen as evidence of AI-generated content. Personally, I’m quite partial to an em dash (you’ll find plenty of them in this newsletter’s back catalog!), but I’ve definitely become more self-conscious about using them in recent months. I find this interesting because it suggests that AI’s use of English, and how we feel about it, will starting shaping our own relationship with the language.

🦸‍♂️ Sorry, Marvel. The Greatest Superhero is Superman

With the exception of Lois & Clark: The New Adventures of Superman, I’ve never much cared for Kal-El. To me, his near-invulnerability and moral infallibility make him less relatable than other superheroes. But Adam Grant would argue that I’ve got it wrong. In this newsletter, Grant makes the case that Superman is not only the most super, but the most human of all the Marvel and DC characters, and references some interesting research about what happens psychologically when kids imagine themselves as superheroes.

👋 And finally…

As a pathological Strava user, this video left me feeling exposed:

@joshpughcomicThanks for the data guys #morningrun #strava #5k #10k #running #runner #joshpugh #gettingitdone

Tiktok failed to load.

Enable 3rd party cookies or use another browser

👍 Thanks!

Thanks for reading The L&D Dispatch from Mind Tools! If you’d like to speak to us, work with us, or make a suggestion, you can email custom@mindtools.com.

Or just hit reply to this email!

Subscribe now

Hey here’s a thing! If you’ve reached all the way to the end of this newsletter, then you must really love it!

Why not share that love by hitting the button below, or just forward it to a friend?

Share